The management of tricuspid valve disease remains controversial and frequently presents a diagnostic and therapeutic challenge for the cardiologist and cardiac surgeon. There is disagreement over whether an abnormal tricuspid valve should be left alone, repaired or replaced, which repair should be performed, and which prosthesis should be implanted. Many patients have concomitant mitral and aortic valve disease, many have had one or more previous operations, and they are often in advanced stages of deterioration. It is, however, possible to improve the results obtained in the treatment of patients with multivalvular heart disease by earlier surgical intervention and by performing a concomitant tricuspid valve procedure, if indicated, before myocardial dysfunction develops.

The tricuspid valve is composed of four anatomical elements: valve leaflets, the annulus, chordae tendineae, and papillary muscles (Fig. 1) 1819. There are three valve leaflets, the anterior leaflet being the largest of the three, and separated from the septal leaflet by a small posterior leaflet. The base of the septal leaflet harbours the penetrating portion of the conducting system. The chordae tendineae of the anterior leaflet are attached to the anterior papillary muscle, which is the dominant papillary muscle in the right ventricle. The posterior and septal leaflets are attached by chordae tendineae to a small posterior papillary muscle and directly to the right ventricular myocardium.

The essential problem in tricuspid stenosis is the inability of the right atrium to empty normally, thus causing the pressure in the right atrium to rise. Tricuspid stenosis is usually of rheumatic origin and occurs in conjunction with mitral and aortic disease, or both. It may also be congenital, or due to carcinoid syndrome or, occasionally, right atrial myxoma.

Stenosis of the tricuspid valve is usually due to diffuse, fibrous thickening of the leaflets, with fusion of two of the three commissures. The chordae tendineae of the tricuspid valve may thicken and shorten, but chordal fusion is generally not as severe as that seen in rheumatic mitral stenosis. The leaflets in tricuspid stenosis usually show minimal calcific deposits, making them amenable to commissurotomy rather than replacement procedures. Unfortunately, most stenotic valves are also moderately regurgitant, therefore usually precluding successful commissurotomy.

Tricuspid stenosis is usually associated with a gradient of at least 5 mmHg, and a valve area that is reduced to less than 1.5 cm². The most frequent symptoms are dyspnoea and fatigue. Elevation of the mean right atrial pressure to above 10 mmHg causes jugular venous distension, hepatosplenomegaly, ascites, and peripheral oedema. The characteristic auscultatory finding is a diastolic rumble, best heard at the left lower sternal border. The influence of respiration helps to differentiate this murmur from that of mitral stenosis: the diastolic rumble of tricuspid stenosis is markedly accentuated during the inspiratory phase. The presence of atrial fibrillation, increased venous pressure, and a diastolic murmur in the classic location should strongly suggest the possibility of significant tricuspid stenosis. The most characteristic radiographic findings include right atrial enlargement, lack of pulmonary artery enlargement, and relatively clear lung fields. M-mode and two-dimensional echocardiography may demonstrate leaflet thickening, reduced diastolic excursion of the leaflets, and characteristic doming of the valve. Doppler echocardiography can allow estimation of the diastolic gradient across the valve.

Tricuspid regurgitation is the result of an incompetent tricuspid valve allowing blood to enter the right atrium during right ventricular contraction. This is most commonly due to enlargement of the annulus secondary to congestive heart failure and pulmonary hypertension. Secondary valvular pathology of this nature is often due to rheumatic heart disease, which in turn causes pulmonary artery hypertension, right ventricular hypertension, and eventually right ventricular dilatation with poor coaptation of the valve leaflets. The most common cause of isolated tricuspid regurgitation is infective endocarditis occurring in intravenous drug abusers. Less common causes of tricuspid regurgitation include trauma, the carcinoid syndrome, and congenital defects such as Ebstein's anomaly.

Symptoms of tricuspid regurgitation include fatigue, dyspnoea, orthopnoea, and peripheral oedema. Distension of the neck veins is common, venous pulsations occurring in a large number of patients. Pulsatile hepatomegaly is usually found as well as the characteristic auscultatory finding of a holosystolic murmur at the left sternal border which increases with respiration. The majority of patients with tricuspid insufficiency have atrial afibrillation and radiographically apparent cardiomegaly. The degree of tricuspid regurgitation can be determined by contrast and/or Doppler echocardiography; two-dimensional echocardiography can be used to evaluate vegetations on the tricuspid valve leaflets, holes in the leaflets, flail leaflets, and abnormal coaptation. Cardiac catheterization reveals the characteristic haemodynamic criteria for tricuspid regurgitation, including a right atrial pressure contour showing ventricularization, (i.e. a prominent V wave with a steep Y descent). In addition, right ventriculography can demonstrate regurgitation of dye into the right atrium and, in severe cases, opacification of the vena cava and hepatic veins.

The decision to proceed with surgery in patients with multivalvular heart disease is usually based on the severity of the aortic and mitral valve disease rather than the severity of tricuspid valve disease. Factors affecting the decision usually include whether a tricuspid valve procedure should be added to the mitral and/or aortic valve procedure and, if so, which tricuspid valve procedure should be performed—annuloplasty or valve replacement. The degree of stenosis and/or regurgitation can be estimated intraoperatively by palpation of the tricuspid valve orifice through the right atrial appendage. If tricuspid valve surgery is not performed as the initial surgical operation, the tricuspid valve orifice can again be examined by digital palpation through the right atrial appendage as bypass is discontinued after the mitral or aortic valve procedure. Intraoperative two-dimensional echocardiography provides precise information regarding the degree of residual valvular insufficiency after repair of either the mitral or tricuspid valve.

Tricuspid stenosis is almost always accompanied by some degree of regurgitation, and can be treated successfully by commissurotomy. Commissurotomy is usually performed under direct vision and at the anteroseptal or posteroseptal commissures, to avoid increasing the degree of tricuspid insufficiency. The procedure may be combined with an annuloplasty ring to correct valve regurgitation. Valve replacement is occasionally necessary, if disease of the leaflets and subvalvular mechanism is advanced or if severe regurgitation cannot be relieved by annuloplasty.

When tricuspid regurgitation occurs secondary to left ventricular failure, dilatation of the annulus occurs chiefly in the right ventricular wall, preventing coaptation of the anterior and posterior leaflets. The relative consistency in size of the septal leaflet allows it to be used as a guide to the size of the tricuspid valve required following reconstruction.

There are three basic reconstructive techniques used in the treatment of tricuspid regurgitation: annular plication, annular ring insertion, and semicircular annuloplasty. Annular plication, first described by Kay in the United States, converts the tricuspid valve into a bicuspid valve by excluding the posterior leaflet, and it is accomplished by plicating the annulus over the posterior leaflet. DeVega described a second type of annuloplasty which consists of a semicircular suture that narrows the annulus along the anterior and posterior leaflets by a purse string suture effect (Fig. 2) 1820. Carpentier devised a semirigid ring with an opening near the area of the septal commissure. His technique distributes the tension over the entire annulus and reduces the incidence of conduction system injury by eliminating sutures in the area of the His bundle (Fig. 3) 1821. Transparent valve obturators are used to size the septal leaflet, which determines the overall size of the tricuspid annulus after a DeVega annuloplasty suture is tied or a Carpentier ring is inserted. Valve competence is checked by filling the right ventricle with cold saline using a bulb syringe. Digital palpation of the tricuspid valve when the patient is no longer maintained on cardiopulmonary bypass can be used to substantiate the effectiveness of the repair and confirm the presence or absence of tricuspid regurgitation. Intraoperative two-dimensional echocardiography may also be used to quantitate any residual tricuspid regurgitation.

The tricuspid valve should be replaced if it is unrepairable (Fig. 4) 1822. When replacing the valve a 4-mm portion of the septal leaflet should be left to avoid injury to the conduction tissue when placing the valve sutures. Placement of the sutures with the heart perfused and beating facilitates recognition of conduction disturbances if the patient is in sinus rhythm and shortens rewarming following associated left-sided cardiac procedures; alternatively, the valve may be replaced under hypothermic cardioplegic arrest. All patients undergoing replacement of the tricuspid valve should have a right ventricular epicardial lead placed at the time of surgery to prevent any conduction abnormalities occurring in the perioperative period. Many surgeons consider the porcine bioprosthesis to be the valve of choice in the tricuspid position because of its associated low incidence of thromboembolic complications, including thrombosis. The St Jude bileaflet mechanical valve, with its low profile and haemodynamics, may offer an advantage in the tricuspid position when compared to other mechanical valves, which tend to have a higher incidence of thrombosis, but it also has a high incidence of thrombosis when compared to tissue valves. The higher incidence of prosthetic thrombosis in the tricuspid position is probably due to the lower pressures involved in the right-sided chambers, lower flow velocities, larger areas of stagnation, and a high incidence of valve entrapment by trabecule.

Right-sided infective endocarditis is one of the most serious complications of intravenous drug abuse, but may also occur following placement of permanent indwelling central venous catheters. The organisms responsible include Staphylococcus aureus, S. epidermidis, Gram-negative bacteria, and fungi. The majority of patients with right-sided endocarditis respond to medical therapy, but some will require surgical treatment for uncontrolled sepsis, recurrent pulmonary emboli, or congestive heart failure. Operative treatment includes vegetectomy, valvuloplasty, valve excision (provided that the pulmonary artery and right ventricular pressures are normal), and valve replacement. Many of these patients are poorly compliant and will return to drug abuse: recurrence of infection is the rule rather than the exception.

Ebstein's anomaly is characterized by downward displacement of the basal attachments of the septal and posterior leaflets of the tricuspid valve. This reduces the chamber size of the right ventricle, a portion of the ventricle above the valve being left as part of the right atrium (the atrialized ventricle). The entire wall of the right ventricle, both above and below the abnormally inserted tricuspid valve, is very thin and dilated. An additional feature is abnormal attachments of the free edge of the valve leaflets, which usually produce a tricuspid valve that is incompetent but may also be stenotic. Disability results from tricuspid valve incompetence and from cyanosis, due to right to left shunting of blood through an associated atrial septal defect or patent foramen ovale. There may be associated anomalies such as pulmonary stenosis or accessory conduction pathways. Surgery is recommended for patients in New York Heart Association class three or four, or those who are severely cyanotic, have paradoxical emboli, or supraventricular and ventricular arrhythmias. Surgical management includes closure of the atrial septal defect or patent foramen ovale, plication of the atrialized portion of the right ventricle, tricuspid valve repair or replacement, correction of any other associated anomalies, and division of any accessory conduction pathways that are present.

Carcinoid heart disease occurs in approximately 50 per cent of patients with carcinoid syndrome and results from metastasis of carcinoid tumours of the midgut. The endocardial myofibromatous lesions usually affect the right side of the heart, causing dysfunction of the tricuspid and pulmonary valves. This may lead to volume and pressure loads on the right-sided cavities and sometimes to right ventricular failure. Valve replacement is recommended before extensive fibrosis and severe right heart failure has developed. Prophylactic somatostatin is administered to block the effects of vasoactive substances that may be released from the tumour during the operative procedure and to reduce the complications caused by their release in the perioperative period.

Mitral valve surgery presents a diagnostic and therapeutic challenge for both the cardiologist and the cardiac surgeon. Recent advances in imaging techniques and a better understanding of the interaction between the mitral valve and the left ventricle have increased the number of patients being referred for reconstruction of the mitral valve. Improvements in myocardial protection and increased experience with valve reconstructive techniques, the availability of second generation bioprosthetic valves, and the newer low profile bileaflet mechanical valves have improved the survival of these patients.

The mitral valve apparatus is a complex co-ordinated mechanism that requires the functional integrity of six anatomical elements working in harmony. These elements are the anterior and posterior valve leaflets, the primary, secondary, and tertiary chordae tendineae, the anterolateral and posteromedial papillary muscles, the annulus, the left ventricular myocardium, and the left atrial endocardium (Fig. 5) 1823. Mounting evidence has indicated that the mitral valve is an integral part of the left ventricle and that its anatomical presence plays a role in left ventricular geometry and mechanics. Papillary and chordal anchoring of the ventricular wall during systole seems essential for efficient left ventricular emptying, and preservation of these structures by either reconstructive surgery or valve replacement seems to be important in reducing the incidence of left ventricular dysfunction in the perioperative period.

With few exceptions, such as a left atrial myxoma, mitral stenosis is always due to chronic rheumatic heart disease. Rheumatic mitral stenosis is characterized pathologically by fibrous fusion of both valve commissures, fibrosis of both leaflets, and fused, thickened, and shortened chordae tendineae, all of which contribute to obstruction of the mitral valve orifice. Annular and subannular calcification, an enlarged left atrium, atrial fibrillation, and thrombus formation can occur as the disease process progresses, unless medical therapy is instituted. The valve area has to reach less than 2 cm² before normal diastolic blood flow begins to be obstructed. Reduction in valve area to 1 cm² is considered to be critical: the left atrial pressures required to achieve adequate ventricular filling through this small orifice will result in appreciable dyspnoea with minimal effort. Patients with mitral stenosis suffer from dyspnoea, fatigue, a gradual reduction in exercise tolerance, cough, haemoptysis, and occasionally angina pectoris, which is related to a very low cardiac output. The classical signs of a stenotic mitral valve include an opening snap, loud first heart sound, and a diastolic rumble. Chest radiographs may show enlargement of the cardiac chambers, especially the left atrium and the right ventricle, and the pulmonary venous pattern may also be pronounced. The ECG will confirm the presence of atrial fibrillation: fibrillation may increase the risk of systemic embolization in patients not receiving anticoagulants. Evaluation of mitral valve disease relies heavily on two-dimensional echocardiography, which provides excellent dynamic images of the leaflets, chordae, papillary muscles, and left ventricular myocardium. It can also define the presence and amount of calcification, valve mobility, and involvement of the subvalvular apparatus, all of which are extremely helpful in deciding whether a stenotic valve is suitable for commissurotomy. Doppler echocardiography can be used to determine the severity of mitral stenosis. Cardiac catheterization is no longer routinely performed, but should be considered in selected patients in whom there are signs or symptoms suggestive of coronary artery disease, or in those over 60 years of age.

Patients with mitral stenosis usually remain asymptomatic for many years, while the severity of mitral stenosis is increasing. It is generally accepted that patients with critical mitral stenosis and severe symptoms of effort dyspnoea, pulmonary oedema, haemoptysis, or right heart failure should undergo surgery. A patient with less severe symptoms might be considered for surgical intervention such as valvuloplasty if he has severe pulmonary hypertension at rest or during exercise, recurrent systemic emboli despite anticoagulant therapy, or reduced left ventricular function which would benefit from relief of mitral valve obstruction by allowing greater left ventricular filling.

Any disruption in the proper systolic closure of the mitral valve may result in mitral regurgitation. Many causes affect different parts of the valvular apparatus. Recent studies indicate that rheumatic heart disease is not the predominant cause of mitral regurgitation. Mitral annular dilatation may be due to chronic left ventricular dilatation, Marfan's syndrome, or secondary subannular calcification. Leaflet pathology, which includes fibrous thickening, calcification, or destruction, can be secondary to rheumatic heart disease, mitral valve prolapse, endocarditis, or Marfan's syndrome. The chordae tendineae may be shortened, thickened, fused, elongated, or may rupture secondary to rheumatic heart disease, mitral valve prolapse, endocarditis, or trauma. Papillary muscle dysfunction, secondary to rupture or fibrosis, can be due to coronary artery disease, myocardial infarction, trauma, or cardiomyopathy. The clinical symptoms and findings in patients with mitral regurgitation can vary, depending upon the severity of mitral regurgitation, its underlying aetiology, whether it is an acute or chronic process, and the status of the left ventricle. Common symptoms include fatiguability, dyspnoea, a decrease in exercise tolerance, chest pain (if coexisting with coronary artery disease), congestive heart failure, and atrial fibrillation as the disease progresses. The classic auscultatory finding for the clinical diagnosis of mitral regurgitation is a holosystolic murmur, with radiation to the axilla and the left infrascapular area. The ECG shows atrial fibrillation in 75 per cent of patients with chronic mitral regurgitation, and the chest radiograph shows cardiac enlargement and pulmonary venous congestion. As mentioned earlier, two-dimensional echocardiography provides excellent dynamic images of the mitral valve apparatus and allows the aetiology of the mitral incompetence to be evaluated, whether it is due to a chordal rupture or elongation, flail leaflets, prolapsing leaflets, calcification, vegetations, or papillary muscle rupture. Doppler echocardiography can be used to determine the severity of mitral regurgitation. Coronary angiography is essential in patients with angina or ventricular dilatation, and in those at risk for coronary artery disease, as well as to confirm estimates of left ventricular function.

All patients with severe mitral regurgitation who are in New York Heart Association class 3, despite good medical management, will benefit from surgical treatment. Many of these patients have a slow undramatic clinical course, which masks the silent but progressive left ventricular dysfunction. By the time the patient is severely symptomatic left ventricular function can have deteriorated markedly, and the results of surgery are far from satisfactory. Patients who have severe mitral regurgitation and who are asymptomatic or minimally symptomatic should undergo non-invasive evaluation of systolic left ventricular function by two-dimensional echocardiography or radionuclide angiography: if the ejection fraction is below 55 per cent surgery is recommended. The patient with severe mitral regurgitation, in whom echocardiography suggests that the valve can be easily repaired, will usually fare better when surgery is undertaken early, before severe symptoms develop, and the potential complications associated with mitral valve replacement at a late stage in the disease can be avoided.

The left atrium can be exposed through a right thoracotomy, a left thoracotomy, a transverse sternotomy, or a median sternotomy. The approach through a right thoracotomy is particularly useful for the small left atrium, for reoperations following previous median sternotomies, and in the presence of patent coronary artery bypass grafts. Left thoracotomy is useful for closed commissurotomy, but is used less today as valve repair gains in popularity. Many surgeons prefer a median sternotomy incision, which provides easy access to most cardiac structures with minimal tissue damage and postoperative pain. Cannulation for cardiopulmonary bypass is accomplished through a single aortic cannula and venous drainage is obtained through cannulation of both cava. The standard approach to the mitral valve is through a left atriotomy, posterior to the intra-atrial groove. If exposure of the mitral valve is not adequate through this incision, the left atriotomy can be extended superiorly and inferiorly around the orifices of the pulmonary veins, the atrium being retracted anteriorly and to the left. Division of the pericardial attachment to the superior and inferior vena cava also permits greater mobility of the heart during anterior retraction. Other approaches include the superior approach through the left atrium, between the superior vena cava laterally and the ascending aorta medially. When a left ventriculotomy is required for resection of a ventricular aneurysm, the mitral valve can also be easily replaced through this incision. In special cases, where the left atrium is particularly small, further exposure can be obtained by the biatrial trans-septal incision of Dubost.

Every effort should be made to repair the mitral valve: this avoids the problems associated with prostheses and leaves an intact mitral valve apparatus, which is important for the satisfactory functioning of the left ventricle. Clinical trials have shown that the results of mitral valve repair are superior to those of mitral valve replacement. The recent development of transoesophageal echocardiography and colour Doppler has aided the intraoperative assessment of the mitral valve. These techniques allow the function of the repaired mitral valve to be assessed in the operating room. The importance of proper exposure of the mitral valve during surgery cannot be overemphasized: special retractors have been developed by Carpentier and others to provide the exposure that is needed. The choice of operative procedure must be based on direct inspection of the mitral valve apparatus by the operating surgeon and the underlying pathology causing the mitral valve dysfunction must be assessed. If the valve cannot be repaired because of severe damage or due to the need for other cardiac procedures such as coronary artery bypass grafting, it may be wise to proceed directly to mitral valve replacement in order to avoid excessive periods of ischaemia.

Closed commissurotomy or digitally dilating a stenotic mitral valve through the left atrial appendage was an important technique in the early days of cardiac surgery, but with the development of safe cardiopulmonary bypass, many surgeons now prefer open mitral commissurotomy and repair of the valve under direct vision. Use of the closed technique is generally restricted to young patients with mobile, non-calcified valves or those with severe stenosis without involvement of the subvalvular apparatus and with no evidence of thrombus, in whom closed commissurotomy by an experienced surgeon can give excellent results.

Open mitral commissurotomy begins with careful inspection of the left atrium and the mitral valve apparatus. If a left atrial thrombus is found it should be removed; the left atrial appendage should be ligated or oversewn as part of the operative procedure. Both commissures should be incised to approximately 4 mm of the mitral valve annulus. In general, patients with more advanced disease, including extensive thickening, shortening, and fusion of the chordae tendineae as well as extensive calcification, would be better served by mitral valve replacement. The operative mortality for mitral commissurotomy is less than 1 per cent. Under normal circumstances, patients may or may not be anticoagulated for the first 3 months following surgery, depending upon the surgeon's preference. However, if a thrombus is found in the left atrium, if the left atrium is extremely large or calcified, or if the patient is in atrial fibrillation, permanent anticoagulation should be instituted.

Mitral valve reconstruction requires a sound knowledge of the functional anatomy of the mitral valve, and each component must be examined individually as the pathology is assessed. The mitral valve must be observed through the opened left atrium while the left ventricle is filled with blood or some other fluid. Injection of saline into the ventricle under pressure will cause the left ventricle to fill and allow the mitral valve leaflets to coapt, thereby allowing physiological assessment of the underlying pathology. Mitral valve reconstructive techniques include chordal shortening, accomplished by involuting the chords into a longitudinal incision in the corresponding papillary muscle; quadrangular resection of the posterior leaflet secondary to chordal rupture, and resuturing the remaining edges of the posterior leaflet (Fig. 6) 1824; and transposition of a quadrangular section of the posterior leaflet to the anterior leaflet, secondary to chordal rupture of the anterior leaflet with resuturing of the posterior leaflet (Fig. 7) 1825. These procedures should all be accompanied by a ring annuloplasty, as performed for patients with annular dilatation. The size of the ring is determined by appropriate sizing of the anterior leaflet. The adequacy of mitral valve repair can be assessed by injecting cold saline through the mitral valve and watching for coaptation of the mitral leaflets or by transoesophageal echocardiography performed as the patient is being weaned from cardiopulmonary bypass. The results of mitral valve reconstruction are satisfactory in terms of both mortality and long-term results.

Mitral valve replacement carries a somewhat greater chance of operative mortality than does mitral valve repair because many of the patients undergoing this operation are old and have associated coronary artery disease and depressed left ventricular function. Preservation of the posterior leaflet and the chordae tendineae has been shown to improve left ventricular function postoperatively, and hopefully prevents the devastating complications of left ventricular rupture which occurs in a small number of patients, especially those who have severe calcification of the mitral valve as well as the subvalvular apparatus. Although artificial heart valves have been available for 25 years there is still no ideal prosthesis, and within rough guidelines applicable to patient age and prosthetic reliability, the choice of valve is generally that of the surgeon. We prefer the St Jude bileaflet mechanical valve (Fig. 8) 1826 or the Carpentier–Edwards porcine bioprosthesis (Fig. 9) 1827 for mitral valve replacement. Both valves function well in the mitral position, have a low incidence of thromboemboli, and have good haemodynamics. The bioprosthetic valve is subject to structural degeneration and, as a rule, requires replacement after 5 to 10 years, especially in patients under the age of 60. The technique of mitral valve replacement is reasonably standardized, with most surgeons using interrupted pledgeted mattress sutures around the mitral annulus (Fig. 10) 1828,1829,1830. It is extremely important that the sutures do not loop around the struts of a bioprosthesis and that redundant leaflet tissue does not interfere with the opening of the low profile St Jude valve. As with all other operations on the left side of the heart, appropriate techniques to remove air must be employed at the end of cardiopulmonary bypass. All patients with mechanical valves should be anticoagulated with coumadin, as should those with biological valves and a large left atrium, atrial afibrillation, or a history of thromboembolism. The results of mitral valve replacement continue to improve due to earlier surgery, better techniques of myocardial preservation, and concomitant repair of associated tricuspid valve and coronary artery disease. Patients with prosthetic mitral valve dysfunction should be operated upon electively once the diagnosis is made: waiting for symptoms to develop is associated with rapidly worsening dysfunction, deterioration of cardiac function, and a marked increase in operative mortality.

Abe T, et al. DeVega's annuloplasty for acquired tricuspid disease. Ann Thoracic Surg 1989; 48: 670–6.

Arbula A, Asfaw I. Tricuspid valvulectomy without prosthetic replacement. J Thoracic Cardiovasc Surg 1981; 82: 684–91.

Assey ME, Spann JE: Indications for heart valve Replacement. Clin Cardiol 1990; 13: 81–8.

Balasundaram SG, Duran C. Surgical approaches to the mitral valve. J Cardiac Surg 1990; 5: 163–9.

Carpentier A, et al. A new reconstructive operation for correction of mitral and tricuspid insufficiency. J Thoracic Cardiovasc Surg 1971; 61: 1–13.

Carpentier A, et al. Surgical management of acquired tricuspid valve disease. J Thoracic Cardiovasc Surg 1974; 67: 53–65.

Carpentier A, et al. Reconstructive surgery of mitral valve incompetence. Ten-year appraisal. J Thoracic Cardiovasc Surg 1980; 79: 338–48.

Cheitlin MD. The timing of surgery in mitral and aortic valve disease. Curr Prob Cardiol 1987; 12: 75–149.

Chopra P, Tandon HD. Pathology of chronic rheumatic heart disease with reference to tricuspid valve involvement. Acta Cardiol 1977; 32: 423–34.

Cobanoglu A, Sturr A. Tricuspid Valve Surgery. In Frankl WS, Brest AN, eds. Valvular heart disease—comprehensive evaluation and management. Philadelphia: Frankl FA Davis Co 1986; 382.

Cosgrove DM, Stewart WJ. Mitral valvuloplasty. Curr Prob Cardiol 1989; 14: 359–415.

Danielson GA. Ebsteiná Anomaly. In: Baue AE, ed. Glenn's thoracic and cardiovascular surgery. Norwalk, Appleton Lange; 1991: 1277.

David TE. A rational approach to the surgical treatment of mitral valve disease. In: Karp REB, Laks H, Wechsler A, eds. Advances in cardiac surgery. Chicago: Mosby Year Book Publishers, 1991: 63–84.

Duran CM. Acquired disease of the mitral valve. In: Baue AE, ed. Glenn's thoracic and cardiovascular surgery. Norwalk: Appleton & Lange, 1991: 1692.

Hammond GL, Geha AS, Kopf GS, Hashim SW. Biological vs. mechanical valves: analyses of 1116 valves inserted in 1012 adult patients with a 4818 patient-year and 5327 valve year follow-up. J Thoracic Cardiovasc Surg 1987; 93: 182–98.

Hauck AJ, et al. Surgical pathology of the tricuspid valve. Mayo Clin Proc 1988; 63: 851–63.

Hughes CF, Noble N. Vegectomy: an alternative surgical treatment for infective endocarditis of the atrioventricular valve in drug addicts. J Thoracic Cardiovasc Surg 1988; 95: 857–61.

Ilbawi MN, et al. Experience with St. Jude medical valve prosthesis in children. J Thoracic Cardiovasc Surg 1987; 93: 73–9.

Karlson KJ, Ashraf MM, Berger RL. Rupture of left ventricle following mitral valve replacement. Ann Thoracic Surg 1988; 46: 590–7.

Kay JH, Mendez AM, Zubiate P. A further look at tricuspid annuloplasty. Ann Thoracic Surg 1976; 22: 498–500.

Lundin L, et al. Surgical treatment of carcinoid heart disease. J Thoracic Cardiovasc Surg 1990; 100: 552–61.

McGrath LB, et al. Tricuspid valve operations in 530 patients. J Thoracic Cardiovasc Surg 1990; 99: 124–33.

Miki, S, et al. Mitral valve replacement with preservations of chordae tendineae and papillary muscles. Ann Thoracic Surg 1988; 45: 28–34.

Mintz GS, et al. Two dimensional echocardiographic evaluation of patients with mitral insufficiency. Am J Cardiol 1979; 44: 670–8.

Miyamura H, Kanazawa H, Hayashi J, Eguchi S. Thombosed St. Jude medical valve prosthesis in the right side of the heart in patients with Tetralogy of Fallot. J Cardiovasc Surg 1987; 94: 148–156.

Morgan JR, et al. Isolated tricuspid stenosis. Circulation 1971; 44: 729–32.

Olson LJ, et al. Surgical pathology of mitral valve. Mayo Clin Proc 1987; 62: 22–34.

O'Rourke RA, Crawford MH. Mitral valve regurgitation. Curr Prob Cardiol 1984; 9: 8–52.

Perfloff JK, Roberts WC. The mitral apparatus: functional anatomy of mitral regurgitation. Circulation 1972; 46: 227–39.

Pupello DF, et al. The Carpentier Edwards bioprosthesis: a comparative study analyzing failure rates by age. J Cardiac Surg 1988; 3: 369–74.

Shimada R, et al. Diagnosis of tricuspid stenosis by M-Mode and two dimensional echocardiography. Am J Cardiol 1984; 53: 164–8.

Singh AK, et al. Follow-up assessment of St. Jude medical prosthetic valve in the tricuspid position: Clinical and hemodynamic results. Ann Thoracic Surg 1984; 37: 324–7.

Salazar E, Levine HD. Rheumatic tricuspid regurgitation. Am J Med 1962; 33: 111–29.

Ubago JL, et al. Right ventriculography as a valid method for the diagnosis of tricuspid insufficiency. Catheterization Cardiovasc Diagnosis 1981; 7: 433–41.

Yee ES, Ullyot DJ. Reparative approach for right-sided endocarditis—operative considerations and results of valvuloplasty. J Cardiovasc Surg 1988; 96: 133.

Zubiate P, Kay JH. Tricuspid stenosis—surgical treatment. Circulation 1964; 29: 95–101.