Valve Repair for Mitral Regurgitation Caused by Isolated Prolapse of the Posterior Leaflet

Transcription

1 Valve Repair for Mitral Regurgitation Caused by Isolated Prolapse of the Posterior Leaflet Patrick Perier, MD, J/2rgen Stumpf, MD, Christian GStz, MD, Fitsoum Lakew, MD, Andr6 Schneider, MD, Bernd Clausnizer, MD, and Robert Hacker, MD Herz und Gef~t~ Klinik, Bad Neustadt/Saale, Germany Background. Although prolapse of the posterior leaflet is the most common abnormality of the mitral valve causing dysfunction, the long-term results of mitral valve repair for this condition are seldom reported. Methods. From October 1988 to June 1994, 208 patients (mean age, 59.4 years) with mitral regurgitation caused by isolated prolapse of the posterior leaflet underwent mitral valve repair alone or combined with myocardial revascularization (n = 30). The surgical techniques were quadrangular resection (n = 199) followed by annulus plication (n = 101) or sliding leaflet plasty (n = 98), use of artificial chordae (n = 5), or papillary muscle shortening (n = 4). All patients had an annuloplasty with a Carpentier ring. Mean follow-up was 3.4 ± 0.1 years and total follow-up, 656 patient-years. Results. There were six operative deaths (2.9%). Postoperative Doppler echocardiography found two cases of systolic anterior motion (1%), and echocardiographic studies at follow-up showed satisfactory mitral valve function in 97% of 112 patients. At 6 years, the actuarial survival rate was 87% + 7%, and freedom from thromboembolic complications, bleeding complications, and reoperation was 93% + 7%, 95% ± 3%, and 95% + 4%, respectively. Conclusions. Mitral valve repair for regurgitation caused by prolapse of the posterior leaflet provides excellent survival at 6 years and should be considered the method of choice for its surgical treatment. (Ann Thorac Surg 1997;64:445-50) 1997 by The Society of Thoracic Surgeons Drolapse of the posterior leaflet as a result of ruptured F chordae was the first abnormality of the regurgitant mitral valve to undergo surgical correction [1]. The term prolapse is broadly used and applied schematically to two different situations. Very often it describes a normally functioning mitral valve with no regurgitation but with a morphologic particularity: the billowing of mitral leaflet bodies toward the left atrium. Conversely, the term prolapse is also used, as in this study, to describe a specific abnormality of a regurgitant mitral valve. The definition of this abnormality is one shared by surgeons [2] and some cardiologists [3]: prolapse of the posterior leaflet of the mitral valve is a failure of leaflet coaptation resulting in displacement of a part of the free edge of the posterior leaflet toward the left atrium. Several surgical techniques have been described to repair prolapse of the posterior leaflet: plication of the flail leaflet segment [1], quadrangular resection followed by either annulus plication [4] or sliding leaflet plasty [5], or artificial chordae implantation [6, 7]. Most often, these techniques are associated with an annuloplasty. The appropriate application of these different techniques is subject to differences of opinion. Although prolapse of the posterior leaflet is the most frequent cause of mitral insufficiency and is easily amenable to repair, follow-up data are few. Here we report our experience with valve repair of isolated prolapse of the posterior leaflet of the Accepted for publication Feb 24, Address reprint requests to Dr Perier, Institut Arnault Tzanck, Avenue Maurice Donat, Saint Laurent du Var, France. mitral valve, present the different surgical techniques performed, and discuss their use. Patients and Methods From October 1988 to June 1994, 522 patients with mitral insufficiency underwent mitral valve repair in our institution. Of this group, 208 (40%) had isolated prolapse of the posterior leaflet, and they form the study cohort. The patients ranged in age from 27 to 79 years (mean age, years). The preoperative clinical profile of these patients is shown in Table 1. Thirty patients had concomitant coronary artery disease, and 12 of them had sustained a myocardial infarction. The preoperative Doppler echocardiographic data obtained in 103 unselected patients are shown in Table 2. Operative Technique A membrane oxygenator with a standard roller pump was used for extracorporeal circulation. Moderate systemic hypothermia (28 C) was employed. Myocardial protection was obtained with a single injection of 500 ml of St. Thomas" cardioplegic solution (formula II) injected into the aortic root and with topical hypothermia. In patients requiring myocardial revascularization, the distal anastomoses were performed prior to mitral valve repair. The proximal anastomoses were completed during the rewarming period. When tricuspid valve repair was indicated, it was performed before mitral valve repair on a fibrillating heart with the aorta unclamped and the left ventricle vented. Cardiopulmonary bypass 1997 by The Society of Thoracic Surgeons /97/$17.00 Published by Elsevier Science Inc PII S (97)

2 446 PERIER ET AL Ann Thorac Surg MITRAL VALVE REPAIR 1997;64: Table 1. Clinical Data a Variable No. of Patients (n = 208) Sex Male 137 (66) Female 71 (34) Electrocardiogram Sinus rhythm 141 (68) Atrial fibrillation 63 (30) Pacemaker 4 (2) New York Heart Association functional class I 19 (9) II 47 (23) III 84 (40) IV 58 (28) Left ventricular ejection fraction b > (70) (13) (3) K (1) Unknown 27 (13) Coronary artery disease 30 (14) One-vessel disease 6 (3) Two-vessel disease 11 (5) Three-vessel disease 13 (6) Associated lesions Tricuspid regurgitation 14 (7) Atrial septal defect 7 (3) Carotid artery stenosis 4 (2) a Numbers in parentheses are percentages, tion was b The mean ejection frac- time and aortic cross-clamp time were 87.8 _ minutes and minutes (mean _+ standard deviation), respectively. Table 3 summarizes the causes and the operative findings. In general repair of the mitral valve was performed by Table 2. Comparison of Preoperative and Postoperative Echocardiographic Data From Subgroup of 103 Patients a P Variable Preop Postop Value Left atrial _+ 9.8 K0.001 diameter (ram) Left ventricular 61.4 _ _+ 7.0 K0.01 end-diastolic diameter (mm) Left ventricular 38.2 _ _+ 8.3 NS end-systolic diameter (mm) Shortening 0.37 _ _ ~0.05 fraction a Data are shown as the mean -+ the standard deviation. NS = not significant. Table 3. Operative Data a Variable No. of Patients Percent Cause Degenerative disease Ischemic mitral regurgitation 9 4 Endocarditis 8 4 Lesions Chordal rupture/elongation Papillary muscle rupture 5 2 Papillary muscle elongation 4 2 Calcified mitral annulus 7 3 Perforation of anterior leaflet 3 1 Annuloplasty ring sizes (mm) techniques described by Carpentier. In 199 patients (96%), the prolapse of the posterior leaflet was corrected with a quadrangular resection of the prolapsed area. The resulting gap in the posterior leaflet was then repaired using two different techniques. The classic method based on a plication of the annulus described by Carpentier and associates [4] was used in 101 patients (51%). In 98 patients (49%), a sliding leaflet plasty was performed [5]. In 5 patients (2%), the prolapse was corrected by chordal replacement with 4-0 polytetrafluoroethylene (Gore-Tex sutures; W.L. Gore & Associates, Flagstaff, AZ) sutures as proposed by David [6] and Frater and colleagues [7]. In 4 patients (2%) with fibrotic elongation of the papillary muscle secondary to coronary heart disease, the prolapse of the posterior leaflet was corrected with a shortening plasty of the papillary muscle. Seven patients had extensive calcification of the posterior portion of the annulus. Mitral valve repair was carried out after decalcification and reconstruction of the annulus as previously described [8]. In all patients, a Carpentier-Edwards annuloplasty ring (model 4400; Baxter Healthcare Corporation, Santa Ana, CA) was used to reduce the size of the annulus, reshape it, and reinforce the repair. The ring sizes used are reported in Table 3. Thirty patients had concomitant myocardial revascularization (2.12 grafts per patient). Two patients experienced acute ischemic mitral regurgitation requiring urgent operation within 2 weeks after a myocardial infarction. Fourteen patients had associated tricuspid valve repair using a Carpentier-Edwards annuloplasty ring. Other associated procedures included closure of an atrial septal defect (7 patients), carotid endarterectomy (4 patients), left ventricular aneurysmectomy (1 patient), and septal myectomy (1 patient). Intraoperative Doppler echocardiography was not used routinely. However, all patients were studied before discharge from the hospital.

3 Ann Thorac Surg PERIER ET AL ;64: MITRAL VALVE REPAIR Anticoa gulation All patients received oral anticoagulants starting 3 days postoperatively (phenprocoumon). The efficacy of anticoagulation was assessed by the international normalized ratio, which was maintained between 3.0 and 4.5. After 3 months, anticoagulant treatment was discontinued at the discretion of the referring physician provided the patient was in sinus rhythm. At the time of the study, 67 patients were anticoagulated, and 32 were receiving antiplatelet therapy. Follow-up Information on hospital mortality and complications was collected as part of the follow-up. Long-term follow-up was completed between January and September 1996 through questionnaires and telephone contacts with patients and the referring physicians. The average duration of follow-up was years. The cumulative follow-up was 656 patient-years. One patient (0.5%) was lost to follow-up. Statistical Analysis Computerized statistical analysis of the data was accomplished using the Sedistat software package (Sedia SA, Paris, France). Standard actuarial and linearized statistical techniques were used to describe survival and incidence of valve-related complications. Student's t test and )(2 contingency tables were used to analyze the significance of differences between preoperative and postoperative data. Continuous data are presented as the mean ± the standard deviation and actuarial probability estimates and linearized rates, as the mean ± two standard errors of the mean. Results were reported according to the recommendations of the Ad Hoc Liaison Committee for Standardizing Definitions of Prosthetic Heart Valve Morbidity [9]. Results Operative Results Six patients died in the hospital, for an operative mortality rate of 2.9%. Of these patients, 5 had had concomitant coronary artery bypass grafting, 2 on an urgent basis within 2 weeks after a myocardial infarction, and 1 had had tricuspid valve repair. The early deaths were due to secondary bleeding complications with tamponade (3 patients), low cardiac output (2 patients), and colon perforation (1 patient). There were no early deaths in the 157 patients with isolated mitral insufficiency. The postoperative complications were as follows: reexploration for bleeding, 8 patients (3.8%); need for a permanent transvenous pacemaker, 7 (3.4%); low cardiac output, 6 (2.9%); ischemic cerebral infarction, 3 (1.4%); transient cerebral ischemic attack, 3 (1.4%); and systolic anterior motion of the mitral leaflet, 2 (1.0%). The mean length of stay in the intensive care unit was 2.1 ± 4.4 days. The patients were discharged from the hospital after 9.2 ± 4.3 days. An echocardiographic evaluation of the repaired mitral valve was performed in 195 patients before discharge. Table 4. Postoperative Echocardiographic Data a'b Variable Value Percent Left atrial diameter (mm) 49.2 _+ 9.4 Left ventricular end-diastolic 52.4 _+ 7.7 diameter (mm) Left ventricular end-systolic 37.7 _+ 8.2 diameter (mm) Shortening fraction 0.28 _ Transvalvular gradient 2.98 _+ 1.2 (ram Hg) Mitral area (cm 2) Mitral regurgitation a Data are shown as the mean -+ the standard deviation for 195 patients. Data on left atrial diameter, left ventricular dimensions, mean gradient across the valve, mitral valve area, and left ventricular shortening fraction are presented in Table 4 together with the evaluation of postoperative mitral regurgitation. Preoperative echocardiographic data obtained in 103 patients were compared with the postoperative data from the same 103 patients (see Table 2). Six to 8 days after mitral valve repair, the left atrial and the left ventricular end-diastolic diameters were significantly reduced. Postoperative echocardiography revealed two cases of systolic anterior motion of the mitral valve with left ventricular outflow tract obstruction. One of these patients had to have mitral valve replacement 1 day after mitral valve repair because of a very unstable hemodynamic condition with low cardiac output. The other patient had a gradient of 40 mm Hg across the left ventricular outflow tract and was successfully treated pharmacologically. Both patients had a quadrangular resection of the posterior leaflet and decalcification of the annulus followed by a sliding leaflet plasty. Survival There have been 13 late deaths. The actuarial survival rates at I year and 6 years were 97% ± 2% and 87% -+ 7% respectively. Figure 1 shows the actuarial survival curve, including early mortality. The causes of late death were sudden death (2 patients), congestive heart failure (3 patients), and noncardiac (8 patients). ThromboemboIic Complications There were eight thromboembolic events, three strokes with residual deficit and five transient cerebral ischemic attacks. Six thromboembolic complications occurred during the postoperative period when anticoagulation was not adequate, and there were two late complications in patients on a regimen of long-term anticoagulation. The size of the implanted ring covered a wide range: 30 in 1 patient, 32 in 3, 34 in 2, and 36 in 2. The freedom from thromboembolic complications was 93% ± 7% at 6 years

4 448 PERIER ET AL Ann Thorac Surg MITRAL VALVE REPAIR 1997;64: z 90 87±7 % z ~, ~ 90 ~ ~ ~ ~ / ~95±3 / 1 :~ 80 ~ 8O ~ I I I [ I I I Fig 1. Actuarial survival. I I I I l I I Fig 3. Actuarial freedom from bleeding complications. 6 7 (Fig 2). The linearized rate of thromboembolic complications was 1.2% +_ 0.4% per patient-year. Anticoagulant-Related Hemorrhage During follow-up, 9 patients sustained an anticoagulantrelated hemorrhage. Three patients had a urogenital hemorrhage, 4 had a gastrointestinal bleeding complication, and 2 had cerebral bleeding leading to death. At 6 years, 95% +- 3% of the patients were free from a bleeding complication (Fig 3). The linearized rate of anticoagulantrelated hemorrhage was 1.4% % per patient-year. Reoperation Eight patients required a reoperation, for an actuarial rate of 95% + 4% freedom from reoperation at 6 years (Fig 4). The reasons for reoperation were recurrent mitral insufficiency caused by chordal rupture (3 patients), ring dehiscence with recurrent mitral regurgitation (2 patients), residual mitral regurgitation (2 patients), and severe systolic anterior motion of the mitral valve (1 patient). Four of the reoperations were performed within the first month after mitral valve repair. A reoperation was required for 4 patients after quadrangular resection and plication of the annulus and for 4 patients after quadrangular resection and sliding leaflet plasty. In all 8 patients, the reoperation was mitral valve replacement. No patient died as a consequence of a reoperation. The linearized rate of reoperation was 1.2% _+ 0.4% per patient-year. Event-Free Survival Combining the incidences of death, thromboembolic complications, reoperation, and anticoagulant-related hemorrhage yielded a composite function describing the event-free survival. Event-free survival at 6 years was 74% _+ 10% (Fig 5). Postoperative Valvular Function Echocardiographic studies at the time of follow-up were available for 112 patients. Eighty-three patients (74.1%) had no mitral regurgitation, 24 (21.4%) had grade I mitral regurgitation, and 5 (4.5%) had grade 2. These results were compared with echocardiographic studies performed on the same 112 patients before discharge. In 109 patients, the function of the mitral valve was considered stable; in only 3 patients was the mitral regurgitation considered more severe. Postoperative Functional Class At the time of follow-up, 130 patients (70%) were in New York Heart Association functional class I, 51 (27%) were in class II, 5 (3%) were in class III, and 1 patient (0.5%) I00 ~ 90 ~7 80 ~o ~ 70 ~e " i i i I T T T i" ! I I I I I Fig 2. Actuarial freedom from thromboembolic complications. z ±7% [ ~ '95±4% 90 8o 70 ~ 6O I I l I I I I Fig 4. Actuarial freedom from reoperation.

5 Ann Thorac Surg PERIER ET AL ;64: MITRAL VALVE REPAIR z c~ Iil Z r~ ~ Fig 5. Actuarial event-free survival. 74±10% was in class IV. We were unable to obtain the postoperative functional class for I patient. One hundred seventysix patients (94%) subjectively thought their condition was improved, 6 (3%) thought it unchanged, and 5 (3%) claimed their condition was worse. Comment Since 19 when McGoon [1] reported the first mitral valve reconstruction, extensive experience with mitral valve repair has been documented [10-12]. These studies present results for nonhomogeneous cohorts of patients with mitral insufficiency resulting from different causes and of different functional types requiring an extremely wide range of surgical techniques. David [12], Cohn [13], and their co-workers reviewed their experience with patients with mitral insufficiency caused only by degenerative disease, but a certain proportion of patients showed preponderant or associated prolapse of the anterior leaflet. The patients in our study exhibited prolapse of only the posterior leaflet, which can be easily repaired using straightforward techniques. Degenerative disease of the mitral valve has been recognized as the most frequent cause of mitral insufficiency in industrialized countries [14]. Analysis of mitral valves with regurgitation resulting from degenerative disease shows that in the majority of cases, the dysfunction is a result of prolapse of the posterior leaflet because of either elongated or ruptured chordae [15]. There is currently little controversy that this lesion should be repaired, as comparative studies [16, 17] demonstrate better long-term survival after valve repair than after valve replacement. It is interesting to note that 9 patients with an ischemic mitral insufficiency had prolapse of the posterior leaflet because of a true ischemic lesion of one head of the posterior papillary muscle supporting chordae of the posterior leaflet exclusively (fibrotic elongation in 5 patients and necrotic rupture in 4). Two of these patients died postoperatively; because of an unstable hemodynamic situation, both had operation on an urgent basis within 2 weeks after a myocardial infarction. Our survival data strongly support the necessity of reconstructing the mitral valve when there is prolapse of the posterior leaflet. The 87% survival rate at 6 years for the entire group of 208 patients is similar to the rates in other reports [10-12]. The actuarial rate of freedom from reoperation of 95% at 6 years compares favorably with results in the literature. Nevertheless, in 3 of our patients, an unsatisfactory repair that led to a reoperation could have been recognized at the time of the first operation if intraoperative transesophageal echocardiography had been routinely performed. This emphasizes the role of intraoperative transesophageal echocardiography to increase the predictability and the stability of mitral valve repair. Most of the thromboembolic complications occurred during the early postoperative period when the patient was not yet anticoagulated. Early anticoagulation with heparin sodium has also been advocated [10] and could be a way to reduce these complications. Nevertheless, we stress that this study is retrospective, the patients and their referring physicians being ask to remember events that occurred several years earlier. The data presented are reasonably accurate for the most dramatic events--death, reoperation, and major bleeding and thromboembolic complications. Minor events such as some thromboembolic or bleeding complications are probably underestimated. Three groups of pathologic features were identified when the mitral valve was examined either at operation or during preoperative echocardiography. These were directly related to the cause and had a direct influence on the choice of surgical technique used. The various techniques at our disposal should not be considered as competitive but rather as alternative possibilities. A very large prolapsed area, which would be impossible to resect as a whole, can be treated by a partial quadrangular resection associated with artificial chordae. A prolapse with localized excess tissue can be corrected with a quadrangular resection followed by a partial slidingleaflet plasty associated with plication of the annulus. It is even possible to combine all of these techniques in a single patient, for instance, quadrangular resection and use of artificial chordae, sliding leaflet plasty, and partial plication of the annulus, if the anatomic and pathologic findings warrant their application. Schematically, the following approach was used to determine which surgical technique to apply: Whenever a floppy or billowing valve was identified, a sliding leaflet plasty was performed [5]. These mitral valves were characterized by an excessive amount of myxomatous tissue in a portion of the posterior mitral leaflet where the distance from its distal margin to its attachment at the mitral annulus was clearly elongated, thus altering the usual 2:1 ratio of the anterior to posterior leaflet surface area. A severe annular dilatation was usually present. As reported by Mihaileanu and associates [18], an excess of valvular tissue is associated with a higher risk of left ventricular outflow tract obstruction. To rain-

6 450 PERIER ET AL Ann Thorac Surg MITRAL VALVE REPAIR 1997;64: imize this risk, the patients with obvious signs of floppy valves had a sliding leaflet plasty to remove this excess tissue, restore a normal ratio of anterior to posterior leaflet surface area, and achieve a regular and progressive narrowing of the mitral annulus, thus avoiding any major change in the geometry of the posterior wall of the left ventricle. The validity of this approach has been presented previously [19]. In our series, this condition was encountered in 98 patients (49%). When the amount of leaflet tissue was not excessive and an associated dilatation of the annulus existed, the repair was achieved by quadrangular resection followed by plication of the annulus and implantation of a ring. This situation corresponds to the fibroelastic deficiency described by Carpentier and colleagues [2]: the leaflets were thin, smooth, and without excess tissue. The annulus was moderately dilated. One hundred one patients (51%) were in this group. When no associated annular dilatation and no excess tissue were detected, a quadrangular resection was avoided, as it would dramatically change the geometry of the posterior leaflet and reduce its functional area. In these patients, the prolapse was corrected by implantation of artificial chordae (5 patients) or shortening of a papillary muscle (4 patients). Whether an annuloplasty ring should be implanted in this group is open to debate. This surgical approach provided good, stable results with only a 1% incidence of systolic anterior motion, which is lower than that in the literature [12, 18, 20]. In summary, the results of mitral valve repair in patients with iscflated prolapse of the posterior leaflet demonstrate low morbidity and mortality, especially when surgical techniques are adapted to the anatomy and the pathology of the mitral apparatus. Mitral valve repair should be considered the method of choice to surgically treat this most common form of mitral insufficiency in industrialized countries. We express our gratitude to Daniela B/iuml and Sylvia Kippnich for their help in the follow-up of patients. References 1. McGoon DC. Repair of mitral valve insufficiency due to ruptured chordae tendineae. J Thorac Cardiovasc Surg 19; 39: Carpentier A, Lessana A, D'Allaines C, Blondeau P, Piwnica A, Dubost C. Reconstruction of mitral valve incompetence. J Thorac Cardiovasc Surg 1980;79: Barlow JB. Aspect of mitral and tricuspid regurgitation. J Cardiol 1991;21: Carpentier A, Relland J, Deloche A, et al. Conservative management of the prolapsed mitral valve. Ann Thorac Surg 1978;26: Carpentier A. The sliding leaflet technique. Le Club Mitrale Newsletter 1988;I David TE. Replacement of chordae tendineae with expanded polytetrafluoroethylene sutures. J Cardiac Surg 1989;4: Frater RWM, Vetter HO, Zussa C, Dahm M. Chordal replacement in mitral valve repair. Circulation 1990;82(Suppl 4): E1 Asmar B, Acker M, Couetil JP, et al. Mitral valve repair in the extensively calcified mitral valve annulus. Ann Thorac Surg 1991;52: Edmunds LH Jr, Clark RE, Cohn LH, Grunkemeier GL, Miller DC, Weisel DR. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1996;62: Deloche A, Jebara VA, Relland JYM, et al. Valve repair with the Carpentier techniques. The second decade. J Thorac Cardiovasc Surg 1990;99: Galloway AC, Colvin SB, Baumann FG, et al. Long-term results of mitral valve reconstruction with Carpentier techniques in 148 patients with mitral valve insufficiency. Circulation 1988;79(Suppl 1): David TE, Armstrong S, Sun Z, Daniel L. Late results of mitral valve repair for mitral regurgitation due to degenerative disease. Ann Thorac Surg 1993;56: Cohn LH, DiSesa VJ, Couper GS, Peigh PS, Kowalker W, Collins JJ. Mitral valve repair for myxomatous degeneration and prolapse of the mitral valve. J Thorac Cardiovasc Surg 1989;98: Waller BF, Morrow AC, Maron BJ, et al. Etiology of clinically isolated chronic pure mitral regurgitation: analysis of 97 patients over 30 years of age having mitral valve replacement. Am Heart J 1982;104: Olson LJ, Subramanian R, Ackermann DM, Orzulak TA, Edwards WD. Surgical pathology of the mitral valve. A study of 712 cases spanning 21 years. Mayo Clin Proc 1987;62: Perier P, Deloche A, Chauvaud S, et al. Comparative evaluation of mitral valve repair and replacement with Starr, Bj6rk and porcine valve prostheses. Circulation 1984;70 (Suppl 1): Akins CW, Hilgenberg AD, Buckley MJ, et al. Mitral valve reconstruction versus replacement for degenerative or ischemic mitral regurgitation. Ann Thorac Surg 1994;58: Mihaileanu S, Marino JP, Chauvaud S, et al. Left ventricular outflow obstruction after mitral valve repair (Carpentier's technique). Circulation 1988;78(Suppl 1): Perier P, Clausnizer B, Mistarz K. Carpentier "sliding leaflet" technique for repair of the mitral valve: early results. Ann Thorac Surg 1994;57: Kreindel MS, Schiavone WA, Lever MH, Cosgrove DM. Systolic anterior motion of the mitral valve after Carpentier ring valvuloplasty for mitral valve prolapse. Am J Cardiol 1986;57: