Midterm results of mitral valve repair with homemade annuloplasty rings1

Midterm Results of Mitral Valve Repair With Homemade Annuloplasty Rings Wiwat Warinsirikul, MD, Pirapat Mokarapong, MD, Surapot Sangchote, MD, Sant Chaiyodsilp, MD, and Supreecha Tanamai, MD Institute of Cardiovascular Diseases, Rajavithi Hospital, Bangkok, Thailand

Background. There are many kinds of prosthetic mitral annuloplasty rings. We report results of our homemade annuloplasty rings. Methods. Between January 1991 and January 1998, 107 patients with mitral insufficiency underwent mitral valve repair with homemade annuloplasty rings. Mitral insufficiency was due to rheumatic disease in 71 patients, degenerative disease in 29, endocarditis in 3, and congenital heart disease in 4 patients. A total of 67 patients were in New York Heart Association functional class III or IV preoperatively. Midterm follow-up was available in 106 patients from 1 month to 6.6 years (average, 2.4 years).

Results. Operative mortality was 0.9%. At 5 years, survival and event-free survival rates were 92% and 80%, and freedom from thromboembolic complications and reoperation were 95% and 93%, respectively. Ninetythree patients (97%) were in New York Heart Association functional class I, 3 patients (3%) were in class II. Echocardiography at follow-up showed satisfactory mitral valve function. Conclusions. Midterm results of homemade annuloplasty rings are comparable to commercial ones. (Ann Thorac Surg 1999;67:63– 6) © 1999 by The Society of Thoracic Surgeons

M

itral valve repair has been used with increasing frequency for the surgical treatment of patients with mitral valve disease. Many kinds of annuloplasty rings and bands are being used [1– 4]. This study describes our experience of midterm results of mitral valve repair with homemade polytetrafluoroethylene (PTFE) annuloplasty rings.

in New York Heart Association functional class I, 33 (30.8%) were in class II, 45 (42.1%) were in class III, and 22 (20.6%) were in class IV. Forty-four patients (41.1%) were in sinus rhythm and 63 patients (58.9%), in atrial fibrillation. All patients had echocardiograms before operation; 75% showed severe mitral regurgitation and 25% moderate regurgitation with associated mitral stenosis.

Patients and Methods

Annuloplasty Rings

Between January 1991 and January 1998, 107 patients with mitral valve insufficiency underwent repair with mitral annuloplasty. There were 44 men and 63 women, age ranged from 3 to 68 years (average, 28 years). Criterion for mitral valve repair in this study was significant mitral regurgitation with progressive left ventricular dilatation. Exclusion criteria were calcified or severe fibrotic valves. Severe shortening and fusion of chordae and papillary muscles, prolapse, or destroyed leaflet that required resection more than one fourth of its length were also excluded. Informed consent was obtained from all patients before participation in this study. This study was approved by the Board of Institute of Cardiovascular Diseases, Rajavithi Hospital on January 3, 1991. The cause of mitral regurgitation was rheumatic heart disease in 71 patients (66.4%), degenerative disease in 29 (27.1%), endocarditis in 3 (2.8%), and congenital heart disease in 4 patients (3.7%). All patients had significant mitral regurgitation. The associated lesions and operative findings are shown in Table 1. Seven patients (6.5%) were

Our rings were constructed with 4-mm diameter PTFE tube and No. 1 stainless-steel wire. A segment of approximately 11-cm long wire was placed inside the lumen of the 10-cm long PTFE tube. The ring was constructed on a cylinder of 3 cm in diameter, as shown in Figure 1. Both ends of the wire were then twisted together making a circular ring of 10 cm in circumference. In such a way this would be approximately 3 cm in diameter and 6.0 to 7.0 cm2 in cross-sectional area. Both ends of the PTFE tube were sewn together with silk. The constructing process was completed. The ring could be reshaped manually to different shapes such as circular, ellipsoid, or asymmetric. Because the framework of the ring was not rigid, its configuration changed during cardiac cycle; therefore, the ring was flexible. The cost of a ring was $150 US. We used the same concepts of mitral valve repair described by Carpentier and associates [1]. Rheumatic patients had more severe fibrotic valves and required more aggressive mobilization technique.

Accepted for publication Jan 16, 1999. Address reprint requests to Dr Warinsirikul, Institute of Cardiovascular Diseases, Rajavithi Hospital, 2 Rajavithi Rd, Bangkok 10400, Thailand.

© 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

This article has been selected for the open discussion forum on the STS Web site: http://www.sts.org/section/atsdiscussion/

0003-4975/99/$20.00 PII S0003-4975(99)00477-4

64

WARINSIRIKUL ET AL HOMEMADE MITRAL ANNULOPLASTY RINGS

Ann Thorac Surg 1999;68:63– 6

Table 1. Operative Findings and Associated Lesions (n ⫽ 107) Variable Annular dilatation Chordal fusion Prolapse Commissural fusion Chordal elongation Chordal rupture Leaflet rupture Tricuspid regurgitation Aortic regurgitation Atrial septal defect

No. of Patients (%) 106 (99.0) 53 (49.5) 38 (35.5) 30 (28.0) 28 (26.2) 16 (15.0) 2 (1.9) 47 (43.9) 29 (27.1) 10 (9.3)

The frequent surgical techniques used for rheumatic heart disease were chordal and papillary muscle splitting; for degenerative disease, leaflet resection and chordal shortening plasty. Our annuloplasty ring was first shaped into shorter (4-cm) anterior portion and a longer (6-cm) posterior portion, then implanted on mitral annulus accordingly with interrupted mattress sutures. The mitral competency was tested after repair by passing a 12F cannula into the left ventricle through cardioplegic hole in the ascending aorta. Water was injected to fill the left ventricle to test the competency of the mitral valve (Fig 2). The shape of the ring was reformed in some ways, for example, change of the curvature, anterior–posterior distance to achieve the best competency. Associated lesions were also repaired. All patients received warfarin for 6 weeks postoperatively. Patients with atrial fibrillation received long-term warfarin. Follow-up is complete in all patients who sur-

Fig 1. The construction of the annuloplasty ring. (A) A segment of approximately 11-cm long wire was placed inside the lumen of the 10-cm long polytetrafluoroethylene tube. The ring was constructed on a cylinder 3 cm in diameter. Both ends of the wire were then twisted together. (B) Both ends of the polytetrafluoroethylene tube were sewn together with silk. (C) The constructing process was completed.

Fig 2. The annuloplasty ring after implantation.

vived the operation from 1 month to 6.6 years, with a total of 256.1 patient-years. Transthoracic echocardiography was performed at 1 week, 1 year, and 5 years after the operation in all patients except those who had reoperations for mitral valve replacement. The ␹2 and Student’s t test were used for statistical analysis of difference between preoperative and postoperative data characteristics. Survival analysis was evaluated by Kaplan-Meier method [5].

Results Hospital mortality was 1 patient (0.9%). One patient was excluded, he required mitral valve replacement because of failure of repair caused by restricted valve mobility. Late deaths occurred in 5 patients. The causes of death ware cerebral hemorrhage due to warfarin overdose (1 patient), cerebral embolism (1), endocarditis (1), and unknown (2 patients). Survival at 5 years was 92% (operative mortality included) (Fig 3). Event-free survival at 5 years was 80%, being better for degenerative disease, which was 92% (operative mortality included) (Fig 4). There were four thromboembolic events in 4 patients for an embolic rate of 1.6% per patient-year. All the events occurred in rheumatic patients who had atrial fibrillation. Freedom from thromboembolic events at 5 years was 95% (Fig 5). Mitral valve replacement was necessary in 4 patients, 2 patients (1.9%) required reoperation within 1 year. Three patients had rheumatic heart disease and the other had degenerative disease. The

Fig 3. Survival after annuloplasty in rheumatic and degenerative disease. (DD ⫽ degenerative disease; N ⫽ number at risk; RHD ⫽ rheumatic heart disease; Total ⫽ all patients.)

Ann Thorac Surg 1999;68:63– 6

WARINSIRIKUL ET AL HOMEMADE MITRAL ANNULOPLASTY RINGS

65

Fig 4. Event-free survival in rheumatic and degenerative disease. (DD ⫽ degenerative disease; N ⫽ number at risk; RHD ⫽ rheumatic heart disease; Total ⫽ all patients.)

Fig 6. Reoperation-free survival in rheumatic and degenerative disease. (DD ⫽ degenerative disease; N ⫽ number at risk; RHD ⫽ rheumatic heart disease; Total ⫽ all patients.)

causes of reoperation were restrictive valve motion in 3 patients and ruptured chordae in 1 patient. Freedom from reoperation was 93% at 5 years (Fig 6). The reoperation rate was 1.6% per patient-year. The patients’ functional class was significantly improved after operation. Ninety-three patients (97%) were in class I and 3 patients (3%) were in class II (p ⬍ 0.01). Results of echocardiography are shown in Table 2 and Figure 7. No systolic anterior motion of the mitral valve was observed.

Its shape can be adjusted manually to achieve the best competence during operation. In our experience, adjustment of ellipsoid shape of the ring gave the best competency. Postoperative echocardiography showed good results of the repair and mitral annulus was restored to normal size and shape. The predictability of the technique is demonstrated by the low incidence of early reoperation (1.9% at 1 year). Prosthetic ring dehiscence has not been seen in our study. Some reports on the use of annuloplasty rings showed incidences of ring dehiscence from negligible to 2.88% [2, 3, 6, 7]. Patients’ survival in this study was 92% at 5 years, which was comparable to other series using commercially available rings [1– 4, 6, 7]. The incidence of reoperation was 1.6% per patient-year with 93% freedom from reoperation at 5 years. Most reoperations were performed in the rheumatic patients because the valves were more fibrotic and the progressive rheumatic process. In our rheumatic patients, we found that there was an increasing degree of regurgitation during follow-up. This study also confirms other reports that the patients with

Comment The function of the annuloplasty ring is to restore dilated mitral annulus. In this study we try to evaluate our homemade PTFE ring. We do not construct the ring according to the length of anterior part of mitral annulus. Diameter of the handmade annuloplasty ring is fixed at 30 mm in the present series based on our experience of mitral valve replacement. The follow-up echocardiography could demonstrate that the orifice was quite adequate. One advantage of this ring is that it is reshapable.

Table 2. Echocardiographic Dataa Postop Variable

Preop (n ⫽ 107)

1 Week (n ⫽ 106)

EDV (mL) ESV (mL) EF (%) MVA-p (mm2) MVA-d (mm2) MPG (mm Hg)

306 ⫾ 153 182 ⫾ 108b 188 ⫾ 107b 182 ⫾ 79b 103 ⫾ 63 95 ⫾ 76 74 ⫾ 62b 69 ⫾ 34b 66 ⫾ 11 49 ⫾ 17b 61 ⫾ 11b 62 ⫾ 10 364 ⫾ 169 286 ⫾ 100b 284 ⫾ 95b 207 ⫾ 41b 241 ⫾ 91 235 ⫾ 68 203 ⫾ 54b 187 ⫾ 52b 8.9 ⫾ 5.6 4.5 ⫾ 1.9b 6.1 ⫾ 2.6b 5.1 ⫾ 1.4b

a Data are means ⫾ standard deviation. tive measurement.

Fig 5. Thromboembolism-free survival in rheumatic and degenerative disease. (DD ⫽ degenerative disease; N ⫽ number at risk; RHD ⫽ rheumatic heart disease; Total ⫽ all patients.)

1 Year (n ⫽ 70)

b

5 Years (n ⫽ 12)

p ⬍ 0.01 versus preopera-

EDV ⫽ end-diastolic volume; EF ⫽ ejection fraction; ESV ⫽ endsystolic volume; MPG ⫽ mean pressure gradient across mitral valve; MVA-d ⫽ mitral valve area (Doppler); MVA-p ⫽ mitral valve area (planimetry); n ⫽ number of patients; Postop ⫽ postoperative measurement; Preop ⫽ preoperatie measurement.

66

WARINSIRIKUL ET AL HOMEMADE MITRAL ANNULOPLASTY RINGS

Ann Thorac Surg 1999;68:63– 6

Limitation of this study is that we could not demonstrate the flexibility of our ring due to the lack of multiplane transesophageal echocardiography in our institute.

We thank our cardiologists: Drs Wilai Puawilai, Sirichai Tanasarnsombat, Saowaluk Prompongsa, Tanarat Choon-ngam, Donpichit Laorakpongse, Poonchai Jitanantwitaya, Sutham Sutheerapatranont, Thanarat Layangool, Chaisit Sangtawesin, and Vachara Jamjureeruk for echocardiography. We thank Dr Suree Athapaisalsarudee, for the original idea, which led to the development of the annuloplasty ring, and Dr Pantpis Sakornpant for his tremendous help and in-depth critique of this study.

Fig 7. Echocardiography showed marked decrease of mitral regurgitation after repair. *p ⬍ 0.01 versus preoperative measurement. (N ⫽ number of patients; Preop ⫽ before operation; 1 week, 1 year, 5 years ⫽ time after operation.)

degenerative disease have a more favorable outcome [2, 6, 8]. Incidence of thromboembolic complications in this study was 1.6% per patient-year. Some reports showed thromboembolic rates of 0.6% to 2.52% per patient-year [1, 9]. Therefore, thromboembolism may not relate to the type of rings. Risk factor for thromboembolism is atrial fibrillation; therefore, the patient who has a risk factor should receive an anticoagulant. We believe that our handmade annuloplasty ring is cheap and gives excellent outcome comparable to the other commercially available rings. The technique is reproducible and this ring is an alternative for using other commercial rings. In summary, this study shows the good midterm results of our handmade annuloplasty ring. The main function of the ring is to restore dilated mitral annulus. It can be made very easily and it is cheap. Long-term follow-up is required to confirm the stability of this report.

References 1. Carpentier A, Chauvaud S, Fabiani JN, et al. Reconstructive surgery of mitral valve incompetence. Ten-year appraisal. J Thorac Cardiovasc Surg 1980;79:338– 48. 2. Deloche A, Jebara VA, Relland JYM, et al. Valve repair with Carpentier techniques. J Thorac Cardiovasc Surg 1990;99:990 – 1002. 3. Cosgrove DM, Chavez AM, Lytle BW, et al. Results of mitral valve reconstruction. Circulation 1986;74(pt 2):I82. 4. Lin FY, Hung, Yang YJ, et al: Mitral valve reconstruction with Carpentier ring for mitral regurgitation: experience with Chinese patients. Taiwan I Hsueh Hui Tsa Chih 1989;88: 330–5. 5. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Statist Assoc 1958;53:457– 81. 6. Galloway AC, Colvin SB, Baumann FG, et al. Long term results of mitral valve reconstruction with Carpentier techniques in 148 patients with mitral insufficiency. Circulation 1988;74(pt2):I97–105. 7. Duran CG, Pomar JL, Revuelta JM, et al. Conservative operation for mitral insufficiency. J Thorac Cardiovasc Surg 1980; 79:326–37. 8. Antunes MJ, Magahaes MP, Colsen PR, et al. Valvuloplasty of rheumatic mitral valve disease. J Thorac Cardiovasc Surg 1987;94:44–56. 9. Duran CG, Revuelta JM, Gait L, et al. Stability of mitral reconstructive surgery at 10-12 years of predominantly rheumatic valvular disease. Circulation 1988;78(Pt2):I91– 6.