- Email: [email protected]
Tricuspid regurgitation in patients with acquired, chronic, pure mitral regurgitation
Volume 94,
Number 4
October 1987
THORACIC AND CARDIOVASCULAR SURGERY The Journal
J
THORAC CARDIOYASC SURG
of
1987;94:481-7
Original Communications
Tricuspid regurgitation in patients with acquired, chronic, pure mitral regurgitation I. Prevalence, diagnosis, and comparison ofpreoperative clinical and hemodynamic features in patients with and without tricuspid regurgitation Most reports of clinical experiences with palliation of acquired tricuspid regurgitation have failed to address the issue of coexisting disease of the mitral or aortic valve, or both. To accurately determine the natural history and the effect of operative interventions, we studied patients with cbonic, pure mitral regurgitation who had surgical treatnlel!t at the National Heart, Lung, and BloodInstitute from 1968 to 1984.Forty-seven patients fulfilled the criteria of a documentedhistory of mitral regurgitation for more tban 1.5 years, minimal mitral diastolic gradient, severe mitral regurgitation by angiography, and no prior mitral or tricuspid operative procedure. Twenty-five of the 47 patients (53%) had evidence of tricuspid regurgitation. No statistical differences in age, sex, mean duration of symptoms of congestive heart failure, or functional class were found between those patients with and those without tricuspid regurgitation. However, patients with symptoms of congestive heart failure for more than 6 years were more fikely to have tricuspid regurgitation. This increased prevalence also correlated with higher elevations of left ventricularend-diastolic,systolic pulmonary artery, and mean right atrial pressures. The severity of tricuspid regurgitation estimated preoperatively did not correlate statistically with that determined by digital palpation, although the presence of tricuspid regurgitation was reliably confirmed. These data demonstrate that tricuspid regurgitation is frequently present in patients with chronic, pure mitral regurgitation andis associated with prolongedsymptoms of congestive heart failure and significant alterations in right heart dynamics.
Steven R. Cohen, MD, Jeffrey E. Sell, MD, Charles L. McIntosh, MD, and Richard E. Clark, MD, Hanover, N.H., and Bethesda. Md. From the Department of Surgery, Dartmouth-Hitchcock Medical Center, Hanover, N.H. (S.R.C.), the Department of Surgery, The Brigham and Women's Hospital, Boston, Mass (J.E.S.), and the Surgery Branch, National Heart, Lung, and Blood Institute, Bethesda, Md. (C.L.M. and R.E.C.).
Received for publication July 16, 1986. Accepted for publication Dec. 23, 1986. Address for reprints: Richard E. Clark, MD, Chief of Surgery, National Heart, Lung, and Blood Institute, 9000 Rockville Pike, Building 10, Rrn 2N-244, Bethesda, MD 20892.
481
The Journal of Thoracic and Cardiovascular Surgery
4 8 2 Cohen et al.
Isolated disease of the tricuspid valve is rare and generally occurs as a result of a congenital malformation such as Ebstein's anomaly. In contrast, acquired disease of the tricuspid valve occurs in a significant proportion of patients with mitral or aortic valve disease, or both.':' It is therefore surprising that the vast majority of reports2, 4, 5. 15 on the management of acquired tricuspid valve disease have focused on the tricuspid valve alone, with little or no attention paid to the type of associated mitral or aortic valve disease. The consequences of mitral regurgitation (MR) on left ventricular function are different from those produced by long-standing mitral stenosis(MS).2, 16, 17 Thus, if the results of various operations on the tricuspid valve are to be assessed, the type of associated mitral valve disease should be addressed. Accordingly, we retrospectively analyzed all patients with mitral valvedisease and tricuspid regurgitation (TR) without aortic valvedisease who were admitted to the Surgery Branch of the National Heart, Lung, and Blood Institute from 1968 to 1984. The patients were separated into one of three groups depending on the nature of their mitral valve lesion:(1) MS and TR, (2) mixed MS and MR and TR, and (3) MR and TR. Our first report deals exclusively with those patients with chronic (> 1.5 years), pure MR with concomitant TR. Part I of the study discusses the incidenceof TR in patients with chronic, pure MR, its diagnosisby conventional techniques, the reliability of digital palpation of the tricuspid valve, and the preoperative clinical and hemodynamic features of patients with and without TR. Part II of the study compares the early and late results in patients undergoing (1) mitral valve replacement alone, (2) concomitant tricuspid valve annuloplasty, and (3) tricuspid valve replacement. Methods Patient population. Clinical records of all patients with mitral valve disease without prior open or closed mitral commissurotomy who underwent either isolated mitral valve replacement or mitral replacement with concomitant tricuspid valve annuloplasty or replacement at the National Heart, Lung, and Blood Institute, Bethesda, Maryland, between Jan. I, 1969, and Jan. I, 1984, were reviewed. There were 61 patients who had predominant MR with or without TR. In four patients the cause of MR was papillary muscle dysfunction owing to ischemic heart disease; in two patients the onset of MR was acute «1,5 years before operation); in two patients MR, although chronic in its course, was a consequence of hypereosinophilia syndrome and complicated by systemic disease; and in an additional six patients interpretation of clinical and hemodynamic findings disclosed late diastolic gradients compatible with MS. These 14 patients, therefore, were eliminated from further study. The remaining 47 patients form the basis of the present study. All fulfilled the
following criteria: (I) a history of MR of more than 1.5 years' duration; (2) an end-diastolic gradient of 4 mm Hg or less across the mitral valve if atrial fibrillation was present at cardiac catheterization; (3) no end-diastolic gradient across the mitral valve if normal sinus rhythm was present at preoperative cardiac catheterization; (4) severe (3+ or 4+/ 4+) MR, as defined in Table I, at preoperative left ventriculography; and (5) no prior history of mitral commissurotomy or other cardiac surgical procedures. TR. TR was assessed in each of the patients preoperatively by clinical history, physical examination, right ventricular catheterization, and/or at operation by digital palpation of the tricuspid valve and visual inspection of the right atrium and tricuspid valve. A clinical history of intermittent or persistent increase in abdominal girth accompanied by pain or pulsations in the right upper quadrant of the abdomen was suggestive of TR. The subjective criteria used to establish the severityof TR by clinical history and physical examination are illustrated in Fig. I. TR was assessed at right heart catheterization, and in the presence of clinical signs, TR was considered to be present if the right atrial pressure was 10 mm Hg or more, with or without large CV waves or confirmation by right ventriculography. All patients underwent mitral valve replacement, and at the time of median sternotomy before the institution of cardiopulmonary bypass, the primary surgeon assessed the presence or absence of TR by digital palpation of the tricuspid valve through the right atrial appendage. The criteria used to establish the severity of TR at operation are illustrated in Fig. 2.
Patient studies. After the establishment of criteria for the diagnosis of MR and TR, the preoperative and postoperative medical and operative records were reviewed for each patient. Operative findings pertaining to the mitral and tricuspid valves were noted. Tricuspid valve structure was graded as normal or abnormal; annular dilatation was assessed and graded normal, moderately dilated, or severely dilated. TR was classified as functional if valve structure was normal and organic if valve structure was abnormal, as determined by the primary surgeon's inspection of the tricuspid valve in the operating room. Statistical methods. Statistical analysis was performed by standard computer-assisted techniques. Life tables and eventfree survival tables were constructed from the National Institutes of Health cardiac data files, The technique used is that of cumulative percentage distribution. IS Patient data were compared on a separately constructed data base. All means are expressed as plus or minus the two-tailed t value times the standard error, which produces a 95% confidence interval." Groups were compared for analysis by the two-tailed unpaired Student's t test. Population characteristics of a nonparametric type were compared by standard x2 analysis. In the case of x2 tables having one or more expected values less than 6, the Fisher exact test was used as a correction for small numbers."
Results Incidence of TR associated with chronic, pure MR. TR was diagnosed in 25 of the 47 patients (53%) on admission to the Surgery Branch of the National
Volume 94 Number 4
Tricuspid regurgitation, I
October 1987
Liver: t Size
SM (1-6) Increases With Inspiration
483
Edema
Tender Pulsatile (j) HJR
Fig. 1. Subjective criteria used to establish the severity of TR on physical examination. Mild = Systolic murmur (SM) increasing with inspiration, mild to moderate jugular venous distention (JVD). and positive hepatojugular reflux (HUR). Moderate = Systolic murmur increasing with inspiration, positive jugular venous distention, and hepatomegaly with or without tenderness and pulsations, remedial to medical therapy. Severe = Systolic murmur increasing with inspiration, ascites, and hepatomegaly with marked pulsations and tenderness, refractory to medical therapy.
Table I. Preoperative clinical characteristics in patients with and without TR Preoperative junctional class (NYHA)
Sex
NaIR TR
Mean age at operation (yr)
No.
%
No.
51 ± 6.39 50 ± 5.49
12 14
55 56
10 II
Male
%
Mean duration of preop. symptoms (yr)
No.
%
No.
%
No.
%
No.
%
45 44
8.2 ± 2.83 13.8 ± 4.53
0 0
0 0
4 3
18 12
II 16
50 64
7 6
32 24
Female
I
Il
III
IV
Legend: NYHA, New York Heart Association.
Heart, Lung, and Blood Institute between January 1969 and August 1982 for operative correction of chronic, pure MR. Twenty of the 47 patients (43%) with chronic, pure MR had no evidence of TR by the aforementioned clinical, hemodynamic, and operative criteria. TR was judged by the senior operating surgeon to be organic or functional, depending on whether tricuspid valve structure was abnormal or normal, respectively. Sufficient information to classify the nature of TR was availablein the operative reports of 16 of the 25 patients (64%) with TR associated with chronic, pure MR. Valve structure was described as normal (functional TR) in 10 of the 16 patients (62%) and abnormal (organic) in six (38%). In only one of the 25 patients (4%) with chronic, pure MR in whom TR was diagnosed preoperatively by clinical history and physical findings was no TR found at operation. It was suspected that TR disappeared as right atrial pressure dropped
precipitously because of hemorrhage during median sternotomy. In fiveof the 25 patients (28%), TR was not suspected preoperatively but was found by digital palpation at operation. No statistical correlation between the degree of TR palpated at operation by digital examination and the severity of TR by history and physical examination was found. Similarly, preoperative hemodynamic data, duration of symptoms of congestive heart failure, patient's gender, and the type of TR (organic versus functional) found at operation had no statistical correlation with the degree of TR assessed by digital palpation. Although not statistically significant, longer duration of preoperative symptoms of congestive heart failure did appear to be associated with more severe degrees of TR. Neither the degree of TR palpated at operation nor the severity of TR by clinical history and physical examination was statistically different between patients with functional TR and those with organic TR. Age at operation,
The Journal of
484
Cohen et al.
Thoracic and Cardiovascular Surgery
Table II. Preoperative cardiac catheterization data
NoTR TR
RA
LVEDP (mm Hg)
PAS (mm Hg)
RVEDP (mmHg)
(mmHg)
12.8 ± 2.42 16.2 ± 2.41
42.9 ± 7.37 65.5 ± 10.16
7.4 ± 1.96 12.2 ± 2.33
7.2 ± 1.7 11.3 ± 2.33
Legend: LVEDP, Left ventricular end-diastolic pressure. PAS, Pulmonary artery systolic pressure. RVEDP, Right ventricular end-diastolic pressure. RA, Mean right atrial pressure.
Fig. 2. Criteria used to quantitate the degree of TR found at operation. 1+ = Small regurgitant jet of TR noted on digital examination through right atrial appendage. 2+ = Obvious regurgitant jet of TR from specific area of valve on digital examiantion. 3+ = Broad front of TR on digital examination. 4+ = Massive TR palpated or seen from right atrial surface and entirely incompetent valve found by digital examination. TV, Tricuspid valve. RV, Right ventricle.
operative catheterization data, and duration of preoperative symptoms were not statistically different in patients with functional TR versus those with organic TR. Preoperative clinical and hemodynamic findings in patients with chronic, pure MR with and without TR. The preoperative clinical characteristics in patients with chronic, pure MR with and without TR are shown in Table I. No statistical differences in age at operation, sex, mean duration of preoperative symptoms of congestive heart failure, or preoperative functional class were found between patients with and without TR. There was a longer duration of preoperative congestive heart failure in patients with TR compared with those with out TR (13.8 ± 4.53 versus 8.2 ± 2.83 years); however, this difference was not statistically significant. Eight embolic events occurred in seven of the 22 patients (32%) without TR, and four embolic events occurred in three of the 25 patients with TR (12%). Thus 10 of the 47 (21%) patients with chronic, pure MR had preoperative thromboembolism. Preoperative physical findings in patients with and without TR are shown in Table I. Patients with TR were more likely to have a systolic murmur increasing with
inspiration, an enlarged pulsatile or tender liver, and positive hepatojugular reflux than patients without TR. However, the presence of ascites, jugular venous, distention, and peripheral edema were not statistically different in patients with or without TR. Similarly, there was no statistically significant difference in the preoperative left ventricular end-diastolic pressure, pulmonary artery pressure, right ventricular end-diastolic pressure, or mean right atrial pressure between patients with TR and those without TR, although pressures were generally higher in patients with TR (Table II). Duration of preoperative symptoms of congestive heart failure. Patients with TR associated with chronic, pure MR had a longer duration of preoperative symptoms of congestive heart failure than patients without TR (14 versus 8 years). Of the 44 patients in whom duration of preoperative symptoms could be determined, 17 (39%) had symptoms for less than 6 years; five patients (29%) had TR and 12 (71%) had no TR. Twenty-seven of the 44 patients (61 %) had duration of preoperative symptoms for more than 6 years; 17 (63%) had TR and only 10 (37%) had no TR. Patients with preoperative symptoms of congestive heart failure for more than 6 years were statistically more likely to have associated TR than patients with preoperative symptoms for less than 6 years (p < 0.05). Left ventricular end-diastolic pressure. Preoperative measurements of left ventricular end-diastolic pressure were available for all patients. In 36 of these 47 patients (77%) this pressure was less than 18 mm Hg and in II (23%) it was greater than 18 mm Hg. TR was present in 16 of the 36 patients (47%) with end-diastolic pressures below 18 mm Hg and in eight of 11 (73%) with pressures above 18 mm Hg. Pulmonary artery systolic pressure. Preoperative measurements of pulmonary artery systolic pressure were available in all 47 patients. In 28% of these patients the pulmonary artery systolic pressure was less than 35 mm Hg and in the other 34 patients (72%) it was greater than 35 mm Hg just before the operation. Of the 13 patients in whom this pressure was less than 35 mm Hg, 10 (77%) had no TR. In contrast, only 12 of the 34 patients (35%) with a pulmonary artery systolic pressure
Volume 94 Number 4 October 1987
of more than 35 mm Hg had no TR and 22 (65%) had TR. TR associated with chronic, pure MR was more likely to be present at operation in patients with a pulmonary pressure greater than 35 mm Hg (p < 0.025). Right ventricular end-diastolic pressure and mean right atrial pressure. Preoperative measurements of right ventricular end-diastolic pressure were available in 46 of the 47 patients (98%) with chronic, pure MR. Only one of the eight patients with a right ventricular end-diastolic pressure higher than 15 mm Hg had no TR as compared with 20 of 38 patients in whom this pressure was less than 15 mm Hg. Preoperative mean right atrial pressures were measured in 44 of the 47 patients (94%) with chronic, pure MR. In 27 of these 44 patients (61%) the mean right atrial pressure was less than 10 mm Hg and in 17 (39%), greater ·than 10 mm Hg. Of the 17 latter patients, only three (18%) had no evidence ofTR and 14 (82%) had TR. Of the 27 patients with a mean right atrial pressure less than 10 mm Hg, 16 (59%) had no TR and 11 (41%) had associated TR. TR was significantly more common in patients with a mean right atrial pressure above 10 mm Hg; conversely, it was significantly less common in patients with a mean right atrial pressure less than 10 mm Hg (p < 0.01). Comments Incidence of TR. Selzer and Katayama" published their classic paper on the clinical patterns, pathophysiology, and natural history of MR in 1972. Although the effects of MR on pulmonary arterial dynamics were discussed, no mention was made of TR or alterations in right ventricular mechanics. Earlier pathologic studies from our institution focused primarily on the causes of pure MR and only briefly mentioned the tricuspid valve." Braunwald' pointed out that left atrial compliance in patients with chronic MR may be (1) normal or reduced, (2) moderately increased, or (3) markedly increased. Although moderately increased left atrial compliance is the most common subgroup, those patients with normal or reduced compliance have little enlargement of the left atrium, but marked elevation of the mean left atrial pressure. Symptoms of pulmonary congestion are common and proliferative changes in the pulmonary arteries may ensue. Consequently, alterations in right ventricular performance occur with hypertrophy and dilation, and eventually the tricuspid valve anulus is affected. Organic involvement of the tricuspid valve may also occur in patients with rheumatic valvular disease independent of hemodynamic factors.' In our population of patients with chronic, pure MR, TR was found in 53% of patients at operation.
Tricuspid regurgitation, I
485
Organic versus functional TR. A variety of disease processes can affect the tricuspid valve apparatus directly and lead to TR. 4 Rheumatic fever is probably the most common cause of organic tricuspid valve disease, and although the incidence of tricuspid valve involvement may be as high as 46%, the pathologic process is usually less severe than that affecting the aortic and mitral valves.' The most common cause of TR, however, is not intrinsic involvement of the valve itself, but dilatation of the right ventricle and of the tricuspid valve ring, which may be complications of right ventricular failure of any cause.' Carpentier and associates" stressed another mechanism that may be responsible for functional TR: excessive dilatation of the right ventricle, which maintains an abnormal tension in the chordae, impairing free motion and preventing perfect coaptation of the leaflets. Anatomic studies from Carpentier's laboratories have shown that dilatation of the anulus does not affect the three tricuspid leaflets to the same degree. The most severely affected is the posterior leaflet; the septal leaflet is almost always unchanged. These studies have given rise to the concept of remodeling the tricuspid anulus by prosthetic rings and further accentuate the need for precise properative characterization of the cause of TR. As Roberts" has pointed out, the origin of MS with or without MR is invariably rheumatic. In contrast, pure MR may result from a variety of causes, including mitral valve prolapse and, less frequently, rheumatic valve disease. No study has reported the relative incidence of organic and functional tricuspid valve disease in patients with chronic, pure MR alone. Organic TR associated with mitral valve disease has been estimated to occur in 17% to 52% of patients, whereas functional TR has been reported to occur in 48% to 83%.4,6,8.10 In our population of patients with pure, chronic MR and associated tricuspid valve disease, organic TR occurred in 35% and functional TR in 59%. Comparison of preopreative clinical and hemodynamic findings in our patients with organic and functional TR was generally unrevealing, although a longer duration of preoperative symptoms of congestive heart failure was noted in patients with abnormal structure of the tricuspid valve. Newer diagnostic techniqus that measure tricuspid annular size and assess dynamic annular function may ultimately be able to differentiate between functional and organic disease of the tricuspid valve and provide useful information about the severity of functional regurgitation before the operation. Clinical and hemodynamic findings. Preoperative clinical and hemodynamic findings in patients with and without TR revealed a trend toward longer duration of
The Journal of
486
Cohen et al.
preoperative symptoms of congestive heart failure. Surprisingly, patients with organic involvement of the tricuspid valve had longer durations of disease than those with functional TR. Perhaps pathologic changes of the tricuspid valve leaflets and anulus eventually develop as consequences of long-standing, functional TR. Patients with chronic, pure MR with higher elevations of left ventricular end-diastolic pressure and those with higher pulmonary artery systolic pressures were more likely to have TR. Seventy-seven percent of patients with pulmonary artery systolic pressures less than 35 mm Hg had no evidence of TR at operation. However, TR was found at surgery in 65% of the patients with pulmonary artery systolic pressures greater than 35 mm Hg. In this study it was unusual to find patients without TR in whom right ventricular end-diastolic pressure was greater than 15 mm Hg. Eighty-two percent of our patients with mean right atrial pressures greater than 10 mm Hg had TR. Diagnosis of TR. Regardless of the underlying mitral valve disease, methods of assessing the severity of TR are needed to consider surgical correction. Angiography is not reliable in grading TR,22.23 and measures such as the use of special catheters and injection of contrast material in small doses and at low velocities, which may improve the diagnostic specificity, are often missed in routine studies.P'" Echocardiography proves useful in diagnosing TR, but attempts to assess TR quantitatively have not been validated." Doppler methods are valuable in semiquantitative diagnosis of TR, and the combined use of a right atrial systolic turbulence scale with analysis of flow-velocity traces from the venous system may improve the noninvasive assessment of the degree of TR. 26 Presently, there is no gold standard for determining the severity of TR. Earlier reports by Breyer and associates' suggested that digital examination of the tricuspid valve before cardiopulmonary bypass was the most accurate method of assessing the severity of TR. Others have disagreed with the reliability of digital palpation, reporting cases in which obvious TR had disappeared at operation as a result of altered circulatory hemodynamics induced by anesthesia or hemorrhage.?" The severity of TR could not be accurately quantified by digital examination of the tricuspid valve; however, we did find it a reliable means of establishing or confirming a diagnosis of TR. Thus, until newer methods of preoperatively quantitating the severity of TR are developed, we agree with Carpentier and colleagues': A constellation of clinical and hemodynamic features, including the individual patient's response to medical therapy, determines the presence of associated TR.
Thoracic and Cardiovascular Surgery
Despite the large number of reports on both MR and TR, little information is available about the prevalence of TR in association with MR or the preoperative clinical and hemodynamic features of patients with and without TR. Moreover, although a significant proportion of patients with severe, acquired mitral valve disease also have evidence of TR, no report has attempted to correlate the prevalence of TR with MR or MS; instead, these distinct groups of patients have been evaluated together as a single entity. It is not clear whether or not the pathogenesis of TR resulting from MR is different from that of TR resulting from MS. It is our contention that whether TR arises because of anatomic destruction of the tricuspid valve or because of right ventricular dilatation with tricuspid annular enlargement, the underlying mitral valve lesion may determine the preoperative and postoperative courses of these patients. Therefore, when evaluating tricuspid valve disease, we urge that patients be categorized by the nature of their underlying mitral and aortic valve lesions. REFERENCES 1. Chopra P, Tandon HD. Pathology of chronic rheumatic heart disease with particular reference to tricuspid valve replacement. Acta Cardiol 1977;32:423-34. 2. Braunwald NS, Ross J Jr, Morrow AG, Conservative management of tricuspid regurgitation in patients undergoing mitral valve replacement. Circulation 1967;35(pt 2):163-9. 3. Braunwald E. Valvular heart disease. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine. Philadelphia: WB Saunders, 1980:1106-27. 4. Khonsari S, Starr A. Acquired disease of the tricuspid valve. In: Sabiston DC, Spencer FC, eds. Gibbon's surgery of the chest, 4th ed. Philadelphia: WB Saunders, 1983:1211-24. 5. 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. 6. Boyd AD, Engelman RM, Isom OW, Reed GE, Spencer Fe. Tricuspid annuloplasty: five and one-half years' experience with 78 patients. J THORAC CARDIOVASC SURG 1974;68:344-51. 7. Grondin P, Meere C, Limet R, Lopez-Bescos L, Delcan JL, Rivera R. Carpentier's annulus and De Vega's annuloplasty: the end of the tricuspid challenge. J THORAC CARDIOVASC SURG 1975;70:852-61. 8. Carpentier A, Deloche A, Hanania G, et al. Surgical management of acquired tricuspid valve disease. J THORAC CARDIOVASC SURG 1974;67:53-65. 9. Duran CMG, Pomar JL, Colman S, Figueoroa A, Revuelta JM, Ubago JL. Is tricuspid valve repair necessary? J THORAC CARDIOVASC SURG 1980;80:849-60.
Volume 94 Number 4 October 1987
10. Sanfelippo PM, Giuliani ER, Danielson GK, Wallace RB, Pluth JR, McGoon DC. Tricuspid valve prosthetic replacement: early and late results with the StarrEdwards prosthesis. J THORAC CARDIOVASC SURG 1976; 71:441-5. II. Kouchoukos NT, Stephenson LW. Indications for and results of tricuspid valve replacement. Adv Cardiol 1976; 17:199-207. 12. Thorburn CW, Morgan JJ, Shanahan MX, Chang YP. Long-term results of tricuspid valve replacement and the problem of prosthetic valve thrombosis. Am J Cardiol 1983;51 :1l28-32. 13. Del Campo C, Akalin H, McKenzie FN. Evaluation of the Carpentier-Edwards porcine bioprosthesis in the tricuspid position. Texas Heart Inst J 1982;9:299-302. 14. Jugdutt BI, Fraser RS, Lee SJK, Rossall RG, Callaghan J'C, Long-term survival after tricuspid valve replacement: results with seven different prostheses. J THORAC CARDlOVASC SURG 1977;74:20-7. 15. Merin G, Borman JB, Stern S, Davidson JT. Surgery for tricuspid valve disease. Isr J Med Sci 1975;11:197-202. 16. Spratt JA, Olsen CO, Tyson GS Jr, et al. Experimental mitral regurgitation: physiological effects of correlation on left ventricular dynamics. J THORAC CARDIOVASC SURG 1983;86:479-489. 17. Phillips HR, Levine FH, Carter JE, et al. Mitral valve replacement for isolated mitral regurgitation: analysis of clinical course and late postoperative left ventricular ejection fraction. Am J Cardiol 1981;48:647-54.
Tricuspid regurgitation, I
487
18. Games PA, Klare GR. Elementary statistics: data analysis for the behavioral sciences, New York: McGraw-Hill Book Company, 1967:56. 19. Schef'ler We. Statistics for health professionals. Reading, Massachusetts: Addison-Wesley, 1984. 20. Selzer A, Katayama F. Mitral regurgitation: clinical patterns, pathophysiology and natural history. Medicine 1972;51:337-66. 21. Roberts WC. Morphologic features of the normal and abnormal mitral valve. Am J Cardiol 1983;51:1005-28. 22. Lingarnneni R, Cha SD, Maranhao Y, Gooch AS, Goldberg H. Tricuspid regurgitation: clinical and angiographic assessment. Cathet Cardiovasc Diagn 1979;5:717. 23. Cairnes KB, Klostor FE, Bristow JD, Lees MH, Griswold HE. Problems in the hemodynamic diagnosis of tricuspid insufficiency. Am Heart J 1968;75:173-9. 24. Miyatake K, Okamoto M, Kinoshita N, et al. Evaluation of tricuspid regurgitation by pulsed Doppler and twodimensional echocardiography. Circulation 1982;66:77784. 25. Garcia-Dorado D, Falzgraf S, Almazan A, Delcan JL, Lopez-Bescos L, Menarguez L. Diagnosis of functional tricuspid insufficiency by pulsed-wave Doppler ultrasound. Circulation 1982;66:1315-21. 26. Pennestri F, Loperfido F, Salvatori MP, et al. Assessment of tricuspid regurgitation by pulsed Doppler ultrasonography of the hepatic veins. Am J Cardiol 1984;54:363-8.
Number 4
October 1987
THORACIC AND CARDIOVASCULAR SURGERY The Journal
J
THORAC CARDIOYASC SURG
of
1987;94:481-7
Original Communications
Tricuspid regurgitation in patients with acquired, chronic, pure mitral regurgitation I. Prevalence, diagnosis, and comparison ofpreoperative clinical and hemodynamic features in patients with and without tricuspid regurgitation Most reports of clinical experiences with palliation of acquired tricuspid regurgitation have failed to address the issue of coexisting disease of the mitral or aortic valve, or both. To accurately determine the natural history and the effect of operative interventions, we studied patients with cbonic, pure mitral regurgitation who had surgical treatnlel!t at the National Heart, Lung, and BloodInstitute from 1968 to 1984.Forty-seven patients fulfilled the criteria of a documentedhistory of mitral regurgitation for more tban 1.5 years, minimal mitral diastolic gradient, severe mitral regurgitation by angiography, and no prior mitral or tricuspid operative procedure. Twenty-five of the 47 patients (53%) had evidence of tricuspid regurgitation. No statistical differences in age, sex, mean duration of symptoms of congestive heart failure, or functional class were found between those patients with and those without tricuspid regurgitation. However, patients with symptoms of congestive heart failure for more than 6 years were more fikely to have tricuspid regurgitation. This increased prevalence also correlated with higher elevations of left ventricularend-diastolic,systolic pulmonary artery, and mean right atrial pressures. The severity of tricuspid regurgitation estimated preoperatively did not correlate statistically with that determined by digital palpation, although the presence of tricuspid regurgitation was reliably confirmed. These data demonstrate that tricuspid regurgitation is frequently present in patients with chronic, pure mitral regurgitation andis associated with prolongedsymptoms of congestive heart failure and significant alterations in right heart dynamics.
Steven R. Cohen, MD, Jeffrey E. Sell, MD, Charles L. McIntosh, MD, and Richard E. Clark, MD, Hanover, N.H., and Bethesda. Md. From the Department of Surgery, Dartmouth-Hitchcock Medical Center, Hanover, N.H. (S.R.C.), the Department of Surgery, The Brigham and Women's Hospital, Boston, Mass (J.E.S.), and the Surgery Branch, National Heart, Lung, and Blood Institute, Bethesda, Md. (C.L.M. and R.E.C.).
Received for publication July 16, 1986. Accepted for publication Dec. 23, 1986. Address for reprints: Richard E. Clark, MD, Chief of Surgery, National Heart, Lung, and Blood Institute, 9000 Rockville Pike, Building 10, Rrn 2N-244, Bethesda, MD 20892.
481
The Journal of Thoracic and Cardiovascular Surgery
4 8 2 Cohen et al.
Isolated disease of the tricuspid valve is rare and generally occurs as a result of a congenital malformation such as Ebstein's anomaly. In contrast, acquired disease of the tricuspid valve occurs in a significant proportion of patients with mitral or aortic valve disease, or both.':' It is therefore surprising that the vast majority of reports2, 4, 5. 15 on the management of acquired tricuspid valve disease have focused on the tricuspid valve alone, with little or no attention paid to the type of associated mitral or aortic valve disease. The consequences of mitral regurgitation (MR) on left ventricular function are different from those produced by long-standing mitral stenosis(MS).2, 16, 17 Thus, if the results of various operations on the tricuspid valve are to be assessed, the type of associated mitral valve disease should be addressed. Accordingly, we retrospectively analyzed all patients with mitral valvedisease and tricuspid regurgitation (TR) without aortic valvedisease who were admitted to the Surgery Branch of the National Heart, Lung, and Blood Institute from 1968 to 1984. The patients were separated into one of three groups depending on the nature of their mitral valve lesion:(1) MS and TR, (2) mixed MS and MR and TR, and (3) MR and TR. Our first report deals exclusively with those patients with chronic (> 1.5 years), pure MR with concomitant TR. Part I of the study discusses the incidenceof TR in patients with chronic, pure MR, its diagnosisby conventional techniques, the reliability of digital palpation of the tricuspid valve, and the preoperative clinical and hemodynamic features of patients with and without TR. Part II of the study compares the early and late results in patients undergoing (1) mitral valve replacement alone, (2) concomitant tricuspid valve annuloplasty, and (3) tricuspid valve replacement. Methods Patient population. Clinical records of all patients with mitral valve disease without prior open or closed mitral commissurotomy who underwent either isolated mitral valve replacement or mitral replacement with concomitant tricuspid valve annuloplasty or replacement at the National Heart, Lung, and Blood Institute, Bethesda, Maryland, between Jan. I, 1969, and Jan. I, 1984, were reviewed. There were 61 patients who had predominant MR with or without TR. In four patients the cause of MR was papillary muscle dysfunction owing to ischemic heart disease; in two patients the onset of MR was acute «1,5 years before operation); in two patients MR, although chronic in its course, was a consequence of hypereosinophilia syndrome and complicated by systemic disease; and in an additional six patients interpretation of clinical and hemodynamic findings disclosed late diastolic gradients compatible with MS. These 14 patients, therefore, were eliminated from further study. The remaining 47 patients form the basis of the present study. All fulfilled the
following criteria: (I) a history of MR of more than 1.5 years' duration; (2) an end-diastolic gradient of 4 mm Hg or less across the mitral valve if atrial fibrillation was present at cardiac catheterization; (3) no end-diastolic gradient across the mitral valve if normal sinus rhythm was present at preoperative cardiac catheterization; (4) severe (3+ or 4+/ 4+) MR, as defined in Table I, at preoperative left ventriculography; and (5) no prior history of mitral commissurotomy or other cardiac surgical procedures. TR. TR was assessed in each of the patients preoperatively by clinical history, physical examination, right ventricular catheterization, and/or at operation by digital palpation of the tricuspid valve and visual inspection of the right atrium and tricuspid valve. A clinical history of intermittent or persistent increase in abdominal girth accompanied by pain or pulsations in the right upper quadrant of the abdomen was suggestive of TR. The subjective criteria used to establish the severityof TR by clinical history and physical examination are illustrated in Fig. I. TR was assessed at right heart catheterization, and in the presence of clinical signs, TR was considered to be present if the right atrial pressure was 10 mm Hg or more, with or without large CV waves or confirmation by right ventriculography. All patients underwent mitral valve replacement, and at the time of median sternotomy before the institution of cardiopulmonary bypass, the primary surgeon assessed the presence or absence of TR by digital palpation of the tricuspid valve through the right atrial appendage. The criteria used to establish the severity of TR at operation are illustrated in Fig. 2.
Patient studies. After the establishment of criteria for the diagnosis of MR and TR, the preoperative and postoperative medical and operative records were reviewed for each patient. Operative findings pertaining to the mitral and tricuspid valves were noted. Tricuspid valve structure was graded as normal or abnormal; annular dilatation was assessed and graded normal, moderately dilated, or severely dilated. TR was classified as functional if valve structure was normal and organic if valve structure was abnormal, as determined by the primary surgeon's inspection of the tricuspid valve in the operating room. Statistical methods. Statistical analysis was performed by standard computer-assisted techniques. Life tables and eventfree survival tables were constructed from the National Institutes of Health cardiac data files, The technique used is that of cumulative percentage distribution. IS Patient data were compared on a separately constructed data base. All means are expressed as plus or minus the two-tailed t value times the standard error, which produces a 95% confidence interval." Groups were compared for analysis by the two-tailed unpaired Student's t test. Population characteristics of a nonparametric type were compared by standard x2 analysis. In the case of x2 tables having one or more expected values less than 6, the Fisher exact test was used as a correction for small numbers."
Results Incidence of TR associated with chronic, pure MR. TR was diagnosed in 25 of the 47 patients (53%) on admission to the Surgery Branch of the National
Volume 94 Number 4
Tricuspid regurgitation, I
October 1987
Liver: t Size
SM (1-6) Increases With Inspiration
483
Edema
Tender Pulsatile (j) HJR
Fig. 1. Subjective criteria used to establish the severity of TR on physical examination. Mild = Systolic murmur (SM) increasing with inspiration, mild to moderate jugular venous distention (JVD). and positive hepatojugular reflux (HUR). Moderate = Systolic murmur increasing with inspiration, positive jugular venous distention, and hepatomegaly with or without tenderness and pulsations, remedial to medical therapy. Severe = Systolic murmur increasing with inspiration, ascites, and hepatomegaly with marked pulsations and tenderness, refractory to medical therapy.
Table I. Preoperative clinical characteristics in patients with and without TR Preoperative junctional class (NYHA)
Sex
NaIR TR
Mean age at operation (yr)
No.
%
No.
51 ± 6.39 50 ± 5.49
12 14
55 56
10 II
Male
%
Mean duration of preop. symptoms (yr)
No.
%
No.
%
No.
%
No.
%
45 44
8.2 ± 2.83 13.8 ± 4.53
0 0
0 0
4 3
18 12
II 16
50 64
7 6
32 24
Female
I
Il
III
IV
Legend: NYHA, New York Heart Association.
Heart, Lung, and Blood Institute between January 1969 and August 1982 for operative correction of chronic, pure MR. Twenty of the 47 patients (43%) with chronic, pure MR had no evidence of TR by the aforementioned clinical, hemodynamic, and operative criteria. TR was judged by the senior operating surgeon to be organic or functional, depending on whether tricuspid valve structure was abnormal or normal, respectively. Sufficient information to classify the nature of TR was availablein the operative reports of 16 of the 25 patients (64%) with TR associated with chronic, pure MR. Valve structure was described as normal (functional TR) in 10 of the 16 patients (62%) and abnormal (organic) in six (38%). In only one of the 25 patients (4%) with chronic, pure MR in whom TR was diagnosed preoperatively by clinical history and physical findings was no TR found at operation. It was suspected that TR disappeared as right atrial pressure dropped
precipitously because of hemorrhage during median sternotomy. In fiveof the 25 patients (28%), TR was not suspected preoperatively but was found by digital palpation at operation. No statistical correlation between the degree of TR palpated at operation by digital examination and the severity of TR by history and physical examination was found. Similarly, preoperative hemodynamic data, duration of symptoms of congestive heart failure, patient's gender, and the type of TR (organic versus functional) found at operation had no statistical correlation with the degree of TR assessed by digital palpation. Although not statistically significant, longer duration of preoperative symptoms of congestive heart failure did appear to be associated with more severe degrees of TR. Neither the degree of TR palpated at operation nor the severity of TR by clinical history and physical examination was statistically different between patients with functional TR and those with organic TR. Age at operation,
The Journal of
484
Cohen et al.
Thoracic and Cardiovascular Surgery
Table II. Preoperative cardiac catheterization data
NoTR TR
RA
LVEDP (mm Hg)
PAS (mm Hg)
RVEDP (mmHg)
(mmHg)
12.8 ± 2.42 16.2 ± 2.41
42.9 ± 7.37 65.5 ± 10.16
7.4 ± 1.96 12.2 ± 2.33
7.2 ± 1.7 11.3 ± 2.33
Legend: LVEDP, Left ventricular end-diastolic pressure. PAS, Pulmonary artery systolic pressure. RVEDP, Right ventricular end-diastolic pressure. RA, Mean right atrial pressure.
Fig. 2. Criteria used to quantitate the degree of TR found at operation. 1+ = Small regurgitant jet of TR noted on digital examination through right atrial appendage. 2+ = Obvious regurgitant jet of TR from specific area of valve on digital examiantion. 3+ = Broad front of TR on digital examination. 4+ = Massive TR palpated or seen from right atrial surface and entirely incompetent valve found by digital examination. TV, Tricuspid valve. RV, Right ventricle.
operative catheterization data, and duration of preoperative symptoms were not statistically different in patients with functional TR versus those with organic TR. Preoperative clinical and hemodynamic findings in patients with chronic, pure MR with and without TR. The preoperative clinical characteristics in patients with chronic, pure MR with and without TR are shown in Table I. No statistical differences in age at operation, sex, mean duration of preoperative symptoms of congestive heart failure, or preoperative functional class were found between patients with and without TR. There was a longer duration of preoperative congestive heart failure in patients with TR compared with those with out TR (13.8 ± 4.53 versus 8.2 ± 2.83 years); however, this difference was not statistically significant. Eight embolic events occurred in seven of the 22 patients (32%) without TR, and four embolic events occurred in three of the 25 patients with TR (12%). Thus 10 of the 47 (21%) patients with chronic, pure MR had preoperative thromboembolism. Preoperative physical findings in patients with and without TR are shown in Table I. Patients with TR were more likely to have a systolic murmur increasing with
inspiration, an enlarged pulsatile or tender liver, and positive hepatojugular reflux than patients without TR. However, the presence of ascites, jugular venous, distention, and peripheral edema were not statistically different in patients with or without TR. Similarly, there was no statistically significant difference in the preoperative left ventricular end-diastolic pressure, pulmonary artery pressure, right ventricular end-diastolic pressure, or mean right atrial pressure between patients with TR and those without TR, although pressures were generally higher in patients with TR (Table II). Duration of preoperative symptoms of congestive heart failure. Patients with TR associated with chronic, pure MR had a longer duration of preoperative symptoms of congestive heart failure than patients without TR (14 versus 8 years). Of the 44 patients in whom duration of preoperative symptoms could be determined, 17 (39%) had symptoms for less than 6 years; five patients (29%) had TR and 12 (71%) had no TR. Twenty-seven of the 44 patients (61 %) had duration of preoperative symptoms for more than 6 years; 17 (63%) had TR and only 10 (37%) had no TR. Patients with preoperative symptoms of congestive heart failure for more than 6 years were statistically more likely to have associated TR than patients with preoperative symptoms for less than 6 years (p < 0.05). Left ventricular end-diastolic pressure. Preoperative measurements of left ventricular end-diastolic pressure were available for all patients. In 36 of these 47 patients (77%) this pressure was less than 18 mm Hg and in II (23%) it was greater than 18 mm Hg. TR was present in 16 of the 36 patients (47%) with end-diastolic pressures below 18 mm Hg and in eight of 11 (73%) with pressures above 18 mm Hg. Pulmonary artery systolic pressure. Preoperative measurements of pulmonary artery systolic pressure were available in all 47 patients. In 28% of these patients the pulmonary artery systolic pressure was less than 35 mm Hg and in the other 34 patients (72%) it was greater than 35 mm Hg just before the operation. Of the 13 patients in whom this pressure was less than 35 mm Hg, 10 (77%) had no TR. In contrast, only 12 of the 34 patients (35%) with a pulmonary artery systolic pressure
Volume 94 Number 4 October 1987
of more than 35 mm Hg had no TR and 22 (65%) had TR. TR associated with chronic, pure MR was more likely to be present at operation in patients with a pulmonary pressure greater than 35 mm Hg (p < 0.025). Right ventricular end-diastolic pressure and mean right atrial pressure. Preoperative measurements of right ventricular end-diastolic pressure were available in 46 of the 47 patients (98%) with chronic, pure MR. Only one of the eight patients with a right ventricular end-diastolic pressure higher than 15 mm Hg had no TR as compared with 20 of 38 patients in whom this pressure was less than 15 mm Hg. Preoperative mean right atrial pressures were measured in 44 of the 47 patients (94%) with chronic, pure MR. In 27 of these 44 patients (61%) the mean right atrial pressure was less than 10 mm Hg and in 17 (39%), greater ·than 10 mm Hg. Of the 17 latter patients, only three (18%) had no evidence ofTR and 14 (82%) had TR. Of the 27 patients with a mean right atrial pressure less than 10 mm Hg, 16 (59%) had no TR and 11 (41%) had associated TR. TR was significantly more common in patients with a mean right atrial pressure above 10 mm Hg; conversely, it was significantly less common in patients with a mean right atrial pressure less than 10 mm Hg (p < 0.01). Comments Incidence of TR. Selzer and Katayama" published their classic paper on the clinical patterns, pathophysiology, and natural history of MR in 1972. Although the effects of MR on pulmonary arterial dynamics were discussed, no mention was made of TR or alterations in right ventricular mechanics. Earlier pathologic studies from our institution focused primarily on the causes of pure MR and only briefly mentioned the tricuspid valve." Braunwald' pointed out that left atrial compliance in patients with chronic MR may be (1) normal or reduced, (2) moderately increased, or (3) markedly increased. Although moderately increased left atrial compliance is the most common subgroup, those patients with normal or reduced compliance have little enlargement of the left atrium, but marked elevation of the mean left atrial pressure. Symptoms of pulmonary congestion are common and proliferative changes in the pulmonary arteries may ensue. Consequently, alterations in right ventricular performance occur with hypertrophy and dilation, and eventually the tricuspid valve anulus is affected. Organic involvement of the tricuspid valve may also occur in patients with rheumatic valvular disease independent of hemodynamic factors.' In our population of patients with chronic, pure MR, TR was found in 53% of patients at operation.
Tricuspid regurgitation, I
485
Organic versus functional TR. A variety of disease processes can affect the tricuspid valve apparatus directly and lead to TR. 4 Rheumatic fever is probably the most common cause of organic tricuspid valve disease, and although the incidence of tricuspid valve involvement may be as high as 46%, the pathologic process is usually less severe than that affecting the aortic and mitral valves.' The most common cause of TR, however, is not intrinsic involvement of the valve itself, but dilatation of the right ventricle and of the tricuspid valve ring, which may be complications of right ventricular failure of any cause.' Carpentier and associates" stressed another mechanism that may be responsible for functional TR: excessive dilatation of the right ventricle, which maintains an abnormal tension in the chordae, impairing free motion and preventing perfect coaptation of the leaflets. Anatomic studies from Carpentier's laboratories have shown that dilatation of the anulus does not affect the three tricuspid leaflets to the same degree. The most severely affected is the posterior leaflet; the septal leaflet is almost always unchanged. These studies have given rise to the concept of remodeling the tricuspid anulus by prosthetic rings and further accentuate the need for precise properative characterization of the cause of TR. As Roberts" has pointed out, the origin of MS with or without MR is invariably rheumatic. In contrast, pure MR may result from a variety of causes, including mitral valve prolapse and, less frequently, rheumatic valve disease. No study has reported the relative incidence of organic and functional tricuspid valve disease in patients with chronic, pure MR alone. Organic TR associated with mitral valve disease has been estimated to occur in 17% to 52% of patients, whereas functional TR has been reported to occur in 48% to 83%.4,6,8.10 In our population of patients with pure, chronic MR and associated tricuspid valve disease, organic TR occurred in 35% and functional TR in 59%. Comparison of preopreative clinical and hemodynamic findings in our patients with organic and functional TR was generally unrevealing, although a longer duration of preoperative symptoms of congestive heart failure was noted in patients with abnormal structure of the tricuspid valve. Newer diagnostic techniqus that measure tricuspid annular size and assess dynamic annular function may ultimately be able to differentiate between functional and organic disease of the tricuspid valve and provide useful information about the severity of functional regurgitation before the operation. Clinical and hemodynamic findings. Preoperative clinical and hemodynamic findings in patients with and without TR revealed a trend toward longer duration of
The Journal of
486
Cohen et al.
preoperative symptoms of congestive heart failure. Surprisingly, patients with organic involvement of the tricuspid valve had longer durations of disease than those with functional TR. Perhaps pathologic changes of the tricuspid valve leaflets and anulus eventually develop as consequences of long-standing, functional TR. Patients with chronic, pure MR with higher elevations of left ventricular end-diastolic pressure and those with higher pulmonary artery systolic pressures were more likely to have TR. Seventy-seven percent of patients with pulmonary artery systolic pressures less than 35 mm Hg had no evidence of TR at operation. However, TR was found at surgery in 65% of the patients with pulmonary artery systolic pressures greater than 35 mm Hg. In this study it was unusual to find patients without TR in whom right ventricular end-diastolic pressure was greater than 15 mm Hg. Eighty-two percent of our patients with mean right atrial pressures greater than 10 mm Hg had TR. Diagnosis of TR. Regardless of the underlying mitral valve disease, methods of assessing the severity of TR are needed to consider surgical correction. Angiography is not reliable in grading TR,22.23 and measures such as the use of special catheters and injection of contrast material in small doses and at low velocities, which may improve the diagnostic specificity, are often missed in routine studies.P'" Echocardiography proves useful in diagnosing TR, but attempts to assess TR quantitatively have not been validated." Doppler methods are valuable in semiquantitative diagnosis of TR, and the combined use of a right atrial systolic turbulence scale with analysis of flow-velocity traces from the venous system may improve the noninvasive assessment of the degree of TR. 26 Presently, there is no gold standard for determining the severity of TR. Earlier reports by Breyer and associates' suggested that digital examination of the tricuspid valve before cardiopulmonary bypass was the most accurate method of assessing the severity of TR. Others have disagreed with the reliability of digital palpation, reporting cases in which obvious TR had disappeared at operation as a result of altered circulatory hemodynamics induced by anesthesia or hemorrhage.?" The severity of TR could not be accurately quantified by digital examination of the tricuspid valve; however, we did find it a reliable means of establishing or confirming a diagnosis of TR. Thus, until newer methods of preoperatively quantitating the severity of TR are developed, we agree with Carpentier and colleagues': A constellation of clinical and hemodynamic features, including the individual patient's response to medical therapy, determines the presence of associated TR.
Thoracic and Cardiovascular Surgery
Despite the large number of reports on both MR and TR, little information is available about the prevalence of TR in association with MR or the preoperative clinical and hemodynamic features of patients with and without TR. Moreover, although a significant proportion of patients with severe, acquired mitral valve disease also have evidence of TR, no report has attempted to correlate the prevalence of TR with MR or MS; instead, these distinct groups of patients have been evaluated together as a single entity. It is not clear whether or not the pathogenesis of TR resulting from MR is different from that of TR resulting from MS. It is our contention that whether TR arises because of anatomic destruction of the tricuspid valve or because of right ventricular dilatation with tricuspid annular enlargement, the underlying mitral valve lesion may determine the preoperative and postoperative courses of these patients. Therefore, when evaluating tricuspid valve disease, we urge that patients be categorized by the nature of their underlying mitral and aortic valve lesions. REFERENCES 1. Chopra P, Tandon HD. Pathology of chronic rheumatic heart disease with particular reference to tricuspid valve replacement. Acta Cardiol 1977;32:423-34. 2. Braunwald NS, Ross J Jr, Morrow AG, Conservative management of tricuspid regurgitation in patients undergoing mitral valve replacement. Circulation 1967;35(pt 2):163-9. 3. Braunwald E. Valvular heart disease. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine. Philadelphia: WB Saunders, 1980:1106-27. 4. Khonsari S, Starr A. Acquired disease of the tricuspid valve. In: Sabiston DC, Spencer FC, eds. Gibbon's surgery of the chest, 4th ed. Philadelphia: WB Saunders, 1983:1211-24. 5. 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. 6. Boyd AD, Engelman RM, Isom OW, Reed GE, Spencer Fe. Tricuspid annuloplasty: five and one-half years' experience with 78 patients. J THORAC CARDIOVASC SURG 1974;68:344-51. 7. Grondin P, Meere C, Limet R, Lopez-Bescos L, Delcan JL, Rivera R. Carpentier's annulus and De Vega's annuloplasty: the end of the tricuspid challenge. J THORAC CARDIOVASC SURG 1975;70:852-61. 8. Carpentier A, Deloche A, Hanania G, et al. Surgical management of acquired tricuspid valve disease. J THORAC CARDIOVASC SURG 1974;67:53-65. 9. Duran CMG, Pomar JL, Colman S, Figueoroa A, Revuelta JM, Ubago JL. Is tricuspid valve repair necessary? J THORAC CARDIOVASC SURG 1980;80:849-60.
Volume 94 Number 4 October 1987
10. Sanfelippo PM, Giuliani ER, Danielson GK, Wallace RB, Pluth JR, McGoon DC. Tricuspid valve prosthetic replacement: early and late results with the StarrEdwards prosthesis. J THORAC CARDIOVASC SURG 1976; 71:441-5. II. Kouchoukos NT, Stephenson LW. Indications for and results of tricuspid valve replacement. Adv Cardiol 1976; 17:199-207. 12. Thorburn CW, Morgan JJ, Shanahan MX, Chang YP. Long-term results of tricuspid valve replacement and the problem of prosthetic valve thrombosis. Am J Cardiol 1983;51 :1l28-32. 13. Del Campo C, Akalin H, McKenzie FN. Evaluation of the Carpentier-Edwards porcine bioprosthesis in the tricuspid position. Texas Heart Inst J 1982;9:299-302. 14. Jugdutt BI, Fraser RS, Lee SJK, Rossall RG, Callaghan J'C, Long-term survival after tricuspid valve replacement: results with seven different prostheses. J THORAC CARDlOVASC SURG 1977;74:20-7. 15. Merin G, Borman JB, Stern S, Davidson JT. Surgery for tricuspid valve disease. Isr J Med Sci 1975;11:197-202. 16. Spratt JA, Olsen CO, Tyson GS Jr, et al. Experimental mitral regurgitation: physiological effects of correlation on left ventricular dynamics. J THORAC CARDIOVASC SURG 1983;86:479-489. 17. Phillips HR, Levine FH, Carter JE, et al. Mitral valve replacement for isolated mitral regurgitation: analysis of clinical course and late postoperative left ventricular ejection fraction. Am J Cardiol 1981;48:647-54.
Tricuspid regurgitation, I
487
18. Games PA, Klare GR. Elementary statistics: data analysis for the behavioral sciences, New York: McGraw-Hill Book Company, 1967:56. 19. Schef'ler We. Statistics for health professionals. Reading, Massachusetts: Addison-Wesley, 1984. 20. Selzer A, Katayama F. Mitral regurgitation: clinical patterns, pathophysiology and natural history. Medicine 1972;51:337-66. 21. Roberts WC. Morphologic features of the normal and abnormal mitral valve. Am J Cardiol 1983;51:1005-28. 22. Lingarnneni R, Cha SD, Maranhao Y, Gooch AS, Goldberg H. Tricuspid regurgitation: clinical and angiographic assessment. Cathet Cardiovasc Diagn 1979;5:717. 23. Cairnes KB, Klostor FE, Bristow JD, Lees MH, Griswold HE. Problems in the hemodynamic diagnosis of tricuspid insufficiency. Am Heart J 1968;75:173-9. 24. Miyatake K, Okamoto M, Kinoshita N, et al. Evaluation of tricuspid regurgitation by pulsed Doppler and twodimensional echocardiography. Circulation 1982;66:77784. 25. Garcia-Dorado D, Falzgraf S, Almazan A, Delcan JL, Lopez-Bescos L, Menarguez L. Diagnosis of functional tricuspid insufficiency by pulsed-wave Doppler ultrasound. Circulation 1982;66:1315-21. 26. Pennestri F, Loperfido F, Salvatori MP, et al. Assessment of tricuspid regurgitation by pulsed Doppler ultrasonography of the hepatic veins. Am J Cardiol 1984;54:363-8.