European Journal of Cardio-thoracic Surgery 23 (2003) 552–559 www.elsevier.com/locate/ejcts
Mitral valve repair for degenerative disease: is pericardial posterior annuloplasty a durable option? Stefano Bevilacquaa,*, Alfredo Giuseppe Cerilloa, Jacopo Gianettia, Umberto Paradossia, Massimiliano Mariania, Sacha Matteuccib, Enkel Kallushia, Mattia Glaubera a
Cardiac Surgery Department, National Research Council, Ospedale G. Pasquinucci CNR-CREAS, Massa, Italy b Cardiovascular Department, Ospedali Riuniti, Bergamo, Italy Received 5 September 2002; received in revised form 16 December 2002; accepted 22 December 2002
Abstract Objective: Biological and prosthestic rings are available for supporting mitral valve repair (MVR). Contrasting data are reported on the durability of pericardial ring annuloplasty. This retrospective study was undertaken to assess the durability of MVR for degenerative regurgitation with posterior annuloplasty performed with glutaraldehyde-treated autologous pericardium. Methods: From August 1995 through December 2000, 133 patients underwent mitral repair for degenerative regurgitation (86 men, age 62.9 ^ 11.5 years). Thirty patients (22.6%) underwent combined coronary artery bypass graft and fourteen (10.5%) underwent tricuspid annuloplasty. Associated aortic disease, previous cardiac surgery and endocarditis were considered exclusion criteria. Results: Seventy-seven patients (57.9%) received a Carpentier – Edwards ring and 56 received (42.1%) an autologous pericardium ring. Thirty-day mortality was 3.8%. Mean follow-up, 98.3% complete, was of 35.6 ^ 18.7 months. Five-year freedom from reoperation and recurrence of mitral regurgitation $ 3 þ /4 þ was significantly higher in the prosthetic ring group (90.1% – CL90%: 81.9 – 98.3%) compared with the pericardial ring group (62.6% – CL90%: 43.1 – 82.1%; P ¼ 0:027). Prosthetic ring implantation (P ¼ 0:004; RR ¼ 0:11) and preoperative New York Heart Association (NYHA) class # II (P ¼ 0:011; RR ¼ 0:16) were independently related to a lower risk of reoperation and recurrence of mitral regurgitation $ 3 þ /4 þ , by multivariate analysis. Five-year overall survival was 91.4% (CL90%: 87.9.7– 95%). A higher preoperative left ventricular end-diastolic diameter (P ¼ 0:006; RR ¼ 1:17) and the severity of associated coronary artery disease (P ¼ 0:021; RR ¼ 2:00) were independent predictive factors for poor survival by multivariate analysis. Conclusions: Posterior pericardial annuloplasty can jeopardize reproducibility and durability of MVR for degenerative regurgitation. q 2003 Published by Elsevier Science B.V. Keywords: Mitral valve repair; Pericardial ring annuloplasty; Reoperation
1. Introduction Mitral valve repair (MVR) is the gold standard for the surgical treatment of degenerative mitral regurgitation. Reparative procedures on the regurgitant mitral valve show better outcomes compared to mitral valve replacement with regard to operative mortality, late survival, and freedom from thromboembolic accidents and endocarditis, with an excellent long-term freedom from reoperation and recurrence of significant regurgitation [1– 8]. The surgical * Corresponding author. Tel.: þ39-0585-4936-04; fax: þ 39-0585-493614. E-mail address:
[email protected] (S. Bevilacqua). 1010-7940/03/$ - see front matter q 2003 Published by Elsevier Science B.V. doi:10.1016/S1010-7940(02)00867-9
technique, as described by Carpentier and colleagues [4], comprises procedures on the valve leaflets, the sub-valvular apparatus and the annulus. Mitral valve annuloplasty with prosthetic ring is generally performed to consolidate the repair, to reshape the annulus and prevent late failure. Posterior pericardial annuloplasty (PPA) has been described as an alternative option, avoiding foreign material implantation and preserving the physiological motion of the mitral annulus [9 –11]. Nevertheless, contrasting data have been reported on the durability of PPA [10 –16]. Therefore, we compared the durability of PPA with prosthetic ring annuloplasty in a group of consecutive patients undergoing MVR for degenerative mitral regurgitation at our institution.
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559
2. Patients and methods 2.1. Patient selection Between August 1995 and December 2000, of 354 patients undergoing mitral valve surgery at our institution, 213 (60.2%) consecutive patients underwent MVR. Of this population, we identified all patients with degenerative mitral regurgitation. Ten cases of failed MVR (4.5%), who underwent mitral valve replacement during the same procedure, were excluded from the study. The etiology of valve lesions was determined according to the conclusive judgment of the operating surgeon, after summarizing the medical history, echocardiographic data and pathologic gross examination at surgery. Mitral regurgitation of rheumatic, endocarditic, congenital or ischemic etiology was considered an exclusion criterion. Associated aortic valve disease and previous cardiac surgery were also exclusion criteria, but combined procedures on the tricuspid valve were not. In the presence of a mitral valve exhibiting evidence of degenerative disease, associated coronary artery disease (CAD) and the need for associated coronary artery bypass grafting (CABG) were not considered exclusion criteria. 2.2. Preoperative assessment The medical charts of the patients fulfilling the inclusion criteria were retrospectively reviewed and processed in a structured database, consisting of demographic data, preoperative clinical data, echocardiographic parameters, operative and postoperative data. All patients received transthoracic echocardiography and cardiac catheterization with selective coronary angiography before the operation. Intra-operative trans-esophageal echocardiography (TEE) was routinely performed to confirm the pathology and assess the repair outcome. 2.3. Surgical procedure Standard moderate hypothermic (28 –328C) cardiopulmonary bypass with aortic and bicaval cannulation was used in all cases, with a conventional approach through a median sternotomy. Antegrade hyperkalemic tepid blood cardioplegia was administered at 15-min intervals. There were only two operating surgeons. The surgical repair was accomplished with the “French correction” technique, as described by Carpentier and colleagues, exposing the valve through a left atriotomy [4]. Chordal replacement with Gore-Tex sutures was not performed in any case. Chordal transposition, chordal shortening by embedding chordae into the papillary muscle (Carpentier technique) or by papillary muscle sliding plasty (Duran technique), were all used to correct anterior leaflet prolapse. Prosthetic ring annuloplasty was performed with Carpentier – Edwards Classic and Carpentier – Edwards Physio ring implantation
553
(Baxter Healthcare Corp., Deerfield, IL, USA). PPA was performed with a gutaraldehyde-treated strip of autologous pericardium, as previously described [9,10]. Just after sternotomy a rectangular patch of autologous pericardium (about 4 £ 5 cm2 ) was harvested, cleaned of adipose tissue and fixed in a 0.65% glutaraldehyde-buffered solution for 15 min. Afterward it was rinsed with saline solution for 15 min in three separate bowls. Just before implantation, a pericardial ring (0.5 – 0.7 cm wide, ‘C’-shaped strip) was cut from the patch. Pericardial ring length was proportioned to the anterior leaflet free edge extension by means of Carpentier obturators. The size of the Carpentier obturator, used for pericardial ring modeling, was properly selected according to patient’s body surface area (BSA). The size was determined according to the following rules: 30 mm obturator for BSA # 1.5 m2; 32 mm obturator for BSA ranging from 1.6 to 1.8 m2; 34 mm obturator for BSA ranging from 1.8 to 2 m2; 36 mm obturator for BSA . 2 m2. The pericardial ring was fixed with mattress sutures along the posterior annulus, just beyond the anatomic commissures. Annuloplasty techniques varied during the study period and in most of cases the choice was not patient-related. Since August 1995 through June 1997, the Carpentier – Edwards Classic ring was the device of choice (20/26 patients – 76.9%). Afterward, from July 1997 to April 1999, the procedure of choice was PPA (47/60 patients – 78.3%). In the last period, from May 1999 through December 2000 we implanted mostly Carpentier – Edwards Physio rings (37/47 patients – 78.7%). There was a significant difference, among the three periods, regarding the distribution of different annuloplasty techniques ðP , 0:0001Þ. 2.4. Postoperative outcomes and follow-up Operative death was defined as any death occurring within 30 days after surgery, without any regard for the place where it occurred. Postoperative complications were defined according to current guidelines [17]. All patients received a transthoracic echocardiography before hospital discharge. Follow-up was conducted during a 2-month interval ending in February 2002 by telephone interview and visit, including physical examination and echocardiography. Deaths attributed to acute myocardial infarction (AMI), congestive heart failure (CHF), arrhythmia and sudden deaths without any other specific cause were considered as cardiovascular deaths. Deaths related to acute endocarditis were also deemed as cardiovascular deaths. Cause of death was established from hospital charts, autopsy reports when available or from a family physician interview. Non-fatal complications such as AMI, stroke, thromboembolism, anticoagulation-related hemorrhage, endocarditis and CHF episodes were all recorded at follow-up.
554
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559
2.5. Echocardiography Two-dimensional and Doppler echocardiography was performed using a Sonos 5500 ultrasound system (Hewlett – Packard). Left ventricular ejection fraction (LVEF) was assessed by the volumetric method [18]. Left ventricular end-diastolic (LVEDD) and end-systolic (LVESD) diameters were calculated by M-mode measurements, guided by 2D images. The grade of mitral regurgitation was evaluated with color-Doppler, using a four-grade, semi-quantitative scale according to the extension of the regurgitant jet in relation to the left atrium [19]. Mitral valve area was calculated by Doppler echocardiography using the pressure half time method. Functional and segmental mitral valve analysis were carried out by intraoperative TEE. The risk of systolic anterior motion (SAM) was estimated on the basis of mitroaortic angle and the ratio of the mitral valve annulus diameter to the anterior mitral leaflets length [20]. TEE was repeated at the end of repair to assess the grade of residual mitral insufficiency and evaluate the occurrence of SAM.
were computed by means of a quadratic approximation to binomial distribution and continuity correction. Overall survival and freedom from reoperation were estimated by Kaplan – Meier analysis. Comparison between unadjusted overall group survival and freedom from reoperation and mitral regurgitation $ 3 þ /4 þ , relative to baseline characteristics, were assessed by the log –rank test. The multivariate relationships of potential predictive factors for late death, for reoperation and mitral regurgitation $ 3 þ /4 þ were evaluated by multivariable Cox regression analysis. Those variables with an univariate P value # 0.1 or those of known biological significance but failing to meet the critical a level were submitted for consideration to multivariable Cox analysis. A stepwise technique was used to enter the selected variables in the analysis. Variables included in the multivariable analysis are listed in Appendix A. Statistical analysis was performed with StatView software version 5.0.1 (SAS Inc., Cary, NC, USA).
3. Results 2.6. Statistical analysis 3.1. Preoperative and operative data Categorical variables were compared between groups by the x 2 test for independence or by the Fisher’s exact test, when appropriated. For the comparison of continuous variables, independent samples t-test was performed when the variable distribution was found to be normal by the Kolmogorov –Smirnov test, otherwise a non-parametric Mann – Whitney test was used. Group data were summarized by mean and standard deviation (SD) or by frequency percentages. Confidence limits of percentages Table 1 Preoperative variablesa
Age, years Male –female BSA (m2) NYHA class (%) Class I Class II Class III Class IV CAD (%) AF (%) Pace-maker (%) LVEF (%) LVEDD (mm) LVESD (mm) LAD (mm) Mitral annulus (mm)
Prosthetic ring
Pericardial ring
P value
61.7 ^ 12.6 47– 30 1.8 ^ 0.17
64.4 ^ 9.7 39– 17 1.8 ^ 0.15
0.182 0.305 0.760 0.189
2 (2.6) 26 (33.8) 46 (59.7) 3 (3.9) 19 (24.7) 23 (29.9) 15 (19.5) 54.5 ^ 10.8 60.4 ^ 5.9 41.3 ^ 5.4 51.3 ^ 8.9 41.4 ^ 5.0
2 (3.6) 27 (48.2) 27 (48.2) 0 11 (19.6) 14 (25) 11 (19.6) 58.5 ^ 9.1 59.8 ^ 5.0 40.8 ^ 5.2 50.8 ^ 7.2 40.1 ^ 4.7
0.493 0.535 0.981 0.024 0.532 0.574 0.719 0.242
a BSA, body surface area; CAD, coronary artery disease; AF, atrial fibrillation, LVEF, left ventricular ejection fraction; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; LAD, left atrium diameter.
One hundred and thirty-three patients (62.4%) showed the eligibility criteria for entering the study. Seventy-seven patients (57.9%) received a prosthetic ring (i.e. 39 Carpentier– Edwards Classic rings and 38 Carpentier – Edwards Physio rings), whereas 56 patients (42.1%) underwent PPA. The two groups were similar with regard to demographics, comorbidities, preoperative clinical status and all preoperative echocardiographic parameters, except LVEF ðP ¼ 0:024Þ (Table 1). Thirty patients (22.6%) underwent combined CABG and 14 (10.5%) underwent tricuspid annuloplasty, with no difference between the two groups. There were no significant differences between the two groups regarding the pathologic anatomy of valve lesions (Table 2). The surgical technique employed for MVR was similar in the two groups, posterior leaflet quadrangular resection with annuloplasty being the most frequent procedure performed. (Table 3). Ring size was significantly larger in the pericardial ring group (Table 3), even if there were no differences in mitral annulus diameter at preoperative echocardiography (Table 1). The difference in ring size between the two groups should be actually intended as the difference between the prosthetic ring size and the size of the Carpentier obturator selected for pericardial ring modeling, and it is not a difference in physical dimension of rings (Table 3). This result must be imputed to the different criteria of size selection between the two groups (as detailed in Section 2). The 30-day mortality was 3.8% (five of 133 patients), with no difference between the two groups (Table 4). All patients died of low cardiac output syndrome. The two groups had a similar postoperative course (Table 4). At
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559 Table 2 Pathologic anatomy
Leaflet prolapse (%) Anterior leaflet Posterior leaflet Both leaflets Annular/leaflet calcification (%)
Prosthetic ring
Pericardial ring
15 (19.5) 27 (35.1) 27 (35.1) 7 (9.1)
10 (17.9) 22 (39.3) 21 (37.5) 4 (7.1)
P value
0.742
0.241
intraoperative TEE, four patients showed SAM with significant left ventricular outflow (LVOT) obstruction after repair. Of note, all these patients had received a Carpentier Classic ring. One of these patients required ring removal during a second period of cardiopulmonary bypass. This patient developed low cardiac output and died 24 h after the operation. In the other cases SAM was successfully treated with volume loading and beta-blockers. Before hospital discharge most of the 30-day survivors showed trivial mitral regurgitation (MR), whereas nine patients (7.0%) had MR ¼ 2 þ =4þ ðmean MR ¼ 0:6 ^ 0:06Þ. Three of these received prosthetic rings (3/47, 4.1%) and six had PPA (6/54, 11.1%) ðP ¼ 0:218Þ. 3.2. Follow-up data Follow-up of 30-day survivors was 98.3% complete at 35.6 ^ 18.7 months (up to 75.9 months). Two patients were lost to follow-up. Total follow-up was 379.7 patients per year. 3.2.1. Durability of the repair Eleven patients required reoperation for recurrent mitral regurgitation at 16.7 ^ 15.6 months, with no operative deaths (prosthetic ring group 2/74 – 2.7%; pericardial ring group 9/54 – 16.7%: P ¼ 0:005). Most of these patients underwent mitral valve replacement but in two cases a rerepair was feasible. Most of repair failures were procedurerelated (Table 5). In the pericardial ring group the most frequent finding at reoperation was suture dehiscence of Table 3 Surgical technique Prosthetic ring Pericardial ring P value Isolated annuloplasty (%) 8 (10.4) Posterior leaflet resection (%) 33 (42.9) Posterior sliding repair (%) 4 (5.2) Chordal transfer (%) 7 (9.1) Chordal shortening (%) 25 (32.5) Carpentier technique 10 (13.0) Duran technique 15 (19.5) Calcium debridement (%) 2 (2.6) 32.0 ^ 2.0 Ring size (mm)a a
See the text for detailed definition.
3 (5.4) 29 (51.8) 6 (10.7) 10 (17.9) 11 (19.6) 2 (3.5) 9 (16.1) 1 (1.8) 33.2 ^ 1.4
0.298 0.308 0.237 0.135 0.100
0.756 0.0004
555
posterior annulus plication and posterior leaflet reconstruction with partial ring detachment (Table 5). Five-year freedom from reoperation was 95% (CL90%: 90.6 –99.4%) for the prosthetic ring group and 79.5% (CL90%: 71.8– 87.2%) for the pericardial ring group ðP ¼ 0:016Þ. Most of the patients showed trivial or mild mitral regurgitation at follow-up. Twenty-two patients (18.5%) had 2 þ /4 þ and four patients (3.4%) had 3 þ /4 mitral regurgitation, with no significant differences between the two groups (prosthetic ring group 2/74 – 2.7%; pericardial ring group 2/54 – 3.7%: P ¼ 0:742). In the prosthetic ring group the two patients showed recurrent prolapse of the anterior leaflet for elongated chordae in one case and chordal rupture in the other one. In the pericardial ring group the mechanism of mitral regurgitation was annular dilatation with lack of leaflet coaptation. These patients have been scheduled for reoperation at our institution. No signs of pericardial ring calcification were found at reoperation or at follow-up echocardiography. Five-year freedom from reoperation and recurrence of mitral regurgitation $ 3 þ /4 þ was significantly higher in the prosthetic ring group (90.1% – CL90%: 81.9 –98.3%) compared with the pericardial ring group (62.6% – CL90%: 43.1 – 82.1%; P ¼ 0:027) (Fig. 1a). Prosthetic ring implantation and preoperative NYHA class # II were independently related to a lower risk of reoperation and recurrence of mitral regurgitation $ 3 þ /4 þ , by multivariate analysis (Table 6). The operating surgeon had no influence on repair durability ðP ¼ 0:706Þ. 3.2.2. Late survival Seven patients died during follow-up. Four deaths were cardiac-related (one patient died of endocarditis), two patients died of stroke and another one of neoplasia. Fiveyear overall survival was 93.3% (CL90%: 90.5 –96.2%). The kind of annuloplasty ring showed no influence on longterm survival (prosthetic ring group: 95.8%; CL90%: 91.8– 99.7% – PPA group: 91.0; CL90%: 83.9 – 98.1%; Table 4 Postoperative dataa
Operative mortality (%) SAM (%) ICU stay (h) MV (h) Re-exploration for bleeding CVA (%) AMI (%) LOS (%) ARF (%) Creatinine .2 mg/dl
Prosthetic ring
Pericardial ring
P value
3 (3.9) 4 (5.2) 29.7 ^ 37.7 14.3 ^ 31.1 2 (2.6) 1 (1.3) 1 (1.3) 5 (6.5) 5 (6.5) 6 (7.8)
2 (3.6) 0 24.8 ^ 19.6 11.1 ^ 20.0 1 (1.8) 0 0 1 (1.8) 3 (5.4) 2 (3.6)
.0.999 0.083 0.381 0.498 .0.999 .0.999 .0.999 0.405 .0.999 0.707
a SAM, systolic anterior motion; ICU, intensive care unit; MV, mechanical ventilation; CVA, cerebro-vascular accident; AMI, acute myocardial infarction; LOS, low output syndrome; ARF, acute respiratory failure.
556
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559
P ¼ 0:519), as depicted in Fig. 1b. The severity of associated CAD (defined by the number of diseased vessels) and higher preoperative LVEDD were independent predictive factors for poor survival by multivariate analysis (Table 7).
Table 5 Causes of failed repair Prosthetic ring Pericardial ring Valve-related, n (%) Progressive disease Endocarditis Procedure-related, n (%) Suture dehiscence Chordal rupture (previously transposed)
1 (1.3) 0
1 (1.8) 0
0 1 (1.3)
5 (9.2) 3 (5.5)
3.2.3. Other outcomes Symptomatic improvement was noted in most of the 30day survivors. The New York Heart Association (NYHA) functional class (1.3 ^ 0.5) at follow-up was significantly better than before the operation (2.3 ^ 0.6), with a mean improvement of 1.3 NYHA class (CL90%: 1.2 –1.4 NYHA class) ðP , 0:0001Þ. There was a significant reduction of LVEDD (mean difference ¼ 7.8 mm – CL90%: 6.9 –8.7 mm; P , 0:0001) and LVESD (mean difference ¼ 7.3 mm – CL90%: 6.4– 8.2 mm; P , 0:0001) at follow-up, compared with preoperative values. On the contrary, LVEF at follow-up (56.1 ^ 8.4%) was unchanged compared with preoperative values. A significant reduction of left atrium dimension was also observed (mean difference ¼ 8.8 mm – CL90%: 7.5– 10.2 mm; P , 0:0001). There were no differences between the groups in terms of LVEDD, LVESD and LVEF variation at follow-up. Mitral valve area was 2.8 ^ 0.3 cm2 with no difference between the two groups. At follow-up no patients showed SAM with significant LVOT obstruction. Thirteen patients among 30-day survivors had non-fatal complications requiring hospitalization during follow-up. One patient (0.8%) had AMI, three patients (2.5%) had a stroke, one patient (0.8%) had peripheral thromboembolism, three patients (2.5%) had anticoagulation-related hemorrhage, and five patients (4.2%) had CHF. Event-free survival at 5 years (freedom from death, reoperation or MR $ 3 þ /4 þ , endocarditis, thromboembolism, stroke, hemorrhage, CHF and AMI) was 60.6% (CL90%: 50.5– 70.8%), with no difference between the two groups (prosthetic ring group: 66.2%; CL90%: 53.8 – 78.7% Table 6 Univariate and multivariate Cox analysis for reoperation and MR $ 3 þ /4 þ
Fig. 1. (a) Freedom from reoperation and mitral regurgitation (MR) $ 3 þ /4 þ . The Kaplan–Meier curve is stratified for the type of implanted ring. P ¼ 0:027. (b) Cumulative survival. The Kaplan–Meier curve is stratified for the type of implanted ring. P ¼ 0:519. (c) Event-free survival (freedom from death, reoperation or MR $ 3 þ /4 þ , endocarditis, thromboembolism, stroke, hemorrhage, CHF and AMI). The Kaplan– Meier curve is stratified for the type of implanted ring. P ¼ 0:484.
Analysis
Relative risk
CL90%
P value
Univariate analysis Preoperative NYHA # II Preoperative AF Prosthetic ring
0.37 0.45 0.29
0.14–0.96 0.19–1.08 0.11–0.77
0.054 0.098 0.026
Multivariate analysis Preoperative NYHA # II Prosthetic ring
0.16 0.11
0.02–0.41 0.03–0.53
0.011 0.004
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559 Table 7 Univariate and multivariate Cox analysis for survivala Analysis
Relative risk
CL90%
P value
Univariate analysis CAD LVEDD LVESD
1.85 1.11 1.11
0.12–3.05 1.00–1.24 1.00–1.23
0.043 0.099 0.099
Multivariate analysis CAD LVEDD
2.00 1.17
1.05–3.33 1.02–1.37
0.021 0.006
a CAD, coronary artery disease (number of diseased vessels); LVEDD, preoperative left ventricular end-diastolic diameter; LVESD, preoperative left ventricular end-systolic diameter.
– PPA group: 50.4%; CL90%: 32.6 –68.1%; P ¼ 0:484), as showed in Fig. 1c.
4. Discussion 4.1. Repair durability In spite of the few available reports considering annular support as a dispensable procedure [21], several authors have demonstrated lack of annuloplasty to jeopardize late results of MVR [1,2]. Annular remodeling and fixation have been regarded as crucial endpoints of the surgical technique in order to assure long-term durability of repair. With the longest follow-up available to date, Carpentier and associates reported a 20-year freedom from reoperation of 92% in all cases annular remodeling with prosthetic ring was performed [2]. Several annuloplasty techniques have been described with the support of both prosthetic and biological devices. Up to date, the ideal annuloplasty ring is a matter of debate. PPA, introduced by Salati and coworkers [9], has been shown to provide more physiological mitral annulus dynamics and to preserve ventricular performance during stress, compared to rigid prosthetic rings [11], with 5-year freedom from reoperation of 89.7%. In a large retrospective study Gillinov and coworkers [1] found that, even if failure to add an annuloplasty to posterior leaflet resection increased the risk of late reoperation, the type of annuloplasty (prosthetic ring or PPA) did not show any impact on long-term durability of repair. Nevertheless, in the same article, it is noteworthy that the Cleveland group, during the last years, did not perform PPA, but preferred prosthetic ring annuloplasty as the procedure of choice [1]. On the other hand, Lorusso and colleagues found that PPA was independently related to late repair failure in patients who underwent ‘double orifice’ MVR [12]. The current study identified PPA as an independent risk factor for MVR failure and reoperation, confirming the results of previous reports [12,13]. In our opinion the use of autologous
557
pericardium does not provide a sufficiently stable and durable fixation and remodeling of the annulus that, by worsening leaflet coaptation, directly contributes to the reoccurrence of mitral regurgitation. Furthermore, as previously demonstrated in an animal model for double orifice repair [22], the absence of a stable annular support could indirectly increase the systolic stress on the repaired structures (transposed or shortened chordae, reconstruction of posterior leaflet resection, etc.), leading to an increased risk of procedure-related failures. This concept, that relates annular diameter and shape with tension on leaflets edges, and consequently, on repair sutures, was confirmed with a mathematical model by Arts et al. [23]. Previous reports by Gillinov et al. [1,24], including also patients treated with PPA, showed that 30– 70% of MVR failures for degenerative disease were procedure-related, with 10 –22% of cases due to suture dehiscence of annuloplasty ring or posterior leaflet repair site. Chauvaud et al. [25] analyzing 73 reoperations after MVR with Carpentier rings, similarly found 11 cases (15.1%) of prosthetic ring dehiscence. In our series nine of 11 late failures (81.9%) were procedurerelated and eight of these (88.9%) were observed in patients who underwent PPA. In the pericardial ring group five of nine (55.6%) reoperated patients showed partial suture dehiscence of ring or posterior leaflet reconstruction, whereas no patients in the prosthetic ring group showed these findings at reoperation. Although the small number of analyzed patients and some particular aspects of our PPA technique could limit the value of our results, it is conceivable to rise some doubts about durability, and especially, reproducibility of PPA. As underlined by Gillinov et al. PPA does not permit to achieve “a measured plication” of the mitral annulus and for this reason, in the last years, they actually abandoned this technique in favor of flexible prosthetic ring annuloplasty [1,24]. The present analysis did not endorse the theoretical advantages of PPA related to the avoidance of foreign material implantation; the type of implanted ring had no impact on endocarditis and thromboembolism incidence. Even if no data are available, it is reasonable that flexible prosthetic rings can offer similar advantages in term of mitral annulus dynamics compared to PPA. Consequently, we have drastically reconsidered the role of PPA in MVR for degenerative disease, moving to Carpentier – Edwards Physio ring as the annular support of choice. 4.2. Late survival Adjunctive results concerning the correlation between higher preoperative NYHA class and increased risk of reoperation, together with the negative prognostic impact of larger preoperative LVEDD on late survival clearly confirm the importance of early indication for surgery in improving long-term outcome of MVR for degenerative disease, as currently recommended. Furthermore, our report showed that the association of CAD has an important impact on
558
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559
long-term survival of patients with degenerative mitral disease. 4.3. Limitations of the study The present study is a retrospective review of the clinical outcome of MVR at a single institution. The kind of implanted annuloplasty ring was not randomly assigned but varied during the study period. Thus, it must be taken into account that biases related to the learning curve could have influenced the results. The etiology of mitral valve disease was not supported by operative specimen histopathology, and this could have introduced confounding heterogeneity of the study population. Furthermore, technical aspects regarding glutaraldehyde fixation (time and concentration), pericardial ring modeling and implantation could be related to the unsatisfactory durability of PPA shown by our results. The follow-up time could be too short and the study population not large enough to reach a conclusive assessment of the two different techniques.
5. Conclusions PPA could negatively affect MVR durability. The present study suggests that PPA could be a technique not reproducible enough to achieve durable annular remodeling and stable fixation for patients with degenerative mitral disease. Larger prospective and randomized studies are needed to confirm these results.
Acknowledgements We are very grateful to Ms Elaine Law and Ms Susan Gwynne for their assistance in manuscript editing.
Appendix A Variables tested in the multivariate Cox analysis. A.1. Preoperative data Age, gender, preoperative NHYA class, hypertension, diabetes, peripheral vascular disease, preoperative serum creatinine $ 2 mg/dl, chronic obstructive pulmonary disease, preoperative atrial fibrillation, chronic heart failure, associated CAD, preoperative LVEF, preoperative left ventricular diameters, preoperative annulus diameter (echocardiographic data), anterior leaflet prolapse, annular calcification. A.2. Operative and postoperative data Operating surgeon, associated tricuspid valve repair,
associated CABG, type of annuloplasty, chordal shortening (Carpentier technique), chordal transposition, papillary muscle sliding plasty, sliding leaflet repair, annular decalcification, mitral regurgitation at hospital discharge.
References [1] Gillinov AM, Cosgrove DM, Blackstone EH, Diaz R, Arnold JH, Lytle BW, Smedira NG, Sabik JF, McCarthy PM, Loop FD. Durability of mitral valve repair for degenerative disease. J Thorac Cardiovasc Surg 1998;116:734–43. [2] Braunberger E, Deloche A, Berrebi A, Abdallah F, Celestin JA, Meimoun P, Chatellier G, Chauvaud S, Fabiani JN, Carpentier A. Very long-term results (more than 20 years) of valve repair with Carpentier’s techniques in nonrheumatic mitral valve insufficiency. Circulation 2001;104(12 Suppl. 1):I8–I11. [3] Deloche A, Jebara VA, Relland JYM, Chauvaud S, Fabiani JN, Perier P, Dreyfus G, Mihaileanu S, Carpentier A. Valve repair with Carpentier techniques: the second decade. J Thorac Cardiovasc Surg 1990;99:990–1002. [4] Carpentier A. Cardiac valve surgery – the “French correction”. J Thorac Cardiovasc Surg 1983;86:323–37. [5] Gillinov AM, Cosgrove DM. Mitral valve repair for degenerative disease. J Heart Valve Dis 2002;11:S15 –S20. [6] Mohty D, Orszulak TA, Schaff HV, Avierinos JF, Tajik JA, Enriquezsarano M. Very long-term survival and durability of mitral valve repair for mitral valve prolapse. Circulation 2001;104:I1–I17. [7] Enriquez-Sarano M, Schaff HV, Orszulak TA, Tajik AJ, Bailey KR, Frye RL. Valve repair improves the outcome of surgery for mitral regurgitation. Circulation 1995;91:1022–8. [8] Akins CW, Hilgenberg AD, Buckley MJ, Vlahakes GJ, Torchiana DF, Daggett WM, Austen WG. Mitral valve reconstruction versus replacement for degenerative or ischemic mitral regurgitation. Ann Thorac Surg 1994;58:668–76. [9] Salati M, Scorfani R, Santoli C. Posterior pericardial annuloplasty: a physiologic correction? Eur J Cardiothorac Surg 1991;5:226–9. [10] Scrofani R, Moriggia S, Salati M, Fundaro P, Danna P, Santoli C. Mitral valve remodeling: long-term results with posterior pericardial annulolasty. Ann Thorac Surg 1996;61:895–9. [11] Borghetti V, Campana M, Scotti C, Domenighini D, Totaro P, Coletti G, Pagani M, Lorusso R. Biological versus prosthetic ring in mitralvalve repair: enhancement of mitral annulus dynamics and leftventricular function with pericardial annuloplasty at long term. Eur J Cardiothorac Surg 2000;17:431–9. [12] Lorusso R, Borghetti V, Totaro P, Parrinello G, Coletti G, Minzioni G. The double-orifice technique for mitral valve reconstruction: predictors of postoperative outcome. Eur J Cardiothorac Surg 2001; 20:583–9. [13] Gillinov AM, Blackstone EH, White J, Howard M, Ahkrass R, Marullo A, Cosrove DM. Durability of combined aortic and mitral valve repair. Ann Thorac Surg 2001;72:20–7. [14] Ng CK, Nesser J, Punzengruber C, Pachinger O, Auer J, Franke H, Hartl P. Valvuloplasty with glutaraldehyde-treated autologous pericardium in patients with complex mitral valve pathology. Ann Thorac Surg 2001;71:78–85. [15] Salvador L, Rocco F, Ius P, Tamari W, Masat M, Paccagnella A, Cesari F, Valfre C. The pericardium reinforced suture annuloplasty: another tool available for mitral annulus repair? J Card Surg 1993;8: 79– 84. [16] Milano A, Codecasa R, De Carlo M, Nardi C, Tartarini G, Verunelli F, Bortolotti U. Mitral valve annuloplasty for degenerative disease: assessment of four different techniques. J Heart Valve Dis 2000;9: 321–6. [17] Edmunds LH, Clark RE, Cohn LH, Miller DC, Starr A. Guidelines for
S. Bevilacqua et al. / European Journal of Cardio-thoracic Surgery 23 (2003) 552–559
[18]
[19]
[20]
[21]
reporting morbidity and mortality after cardiac valvular operations. J Thorac Cardiovasc Surg 1996;112:708– 11. Naik MM, Diamond GA, Pai T, Soffer A, Siegel RJ. Correspondence of left ventricular ejection fraction determinations from twodimensional echocardiography, radionuclide angiography and contrast cineangiography. J Am Coll Cardiol 1995;25:937–42. Helmcke F, Nanda NC, Hsiung MC, Soto B, Adey CK, Goyal RG, Gatewood Jr RP. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175 –83. Maslow AD, Regan MM, Haering JM, Johnson RG, Levine RA. Echocardiographic predictors of left ventricular outflow tract obstruction and systolic anterior motion of the mitral valve after mitral valve reconstruction for myxomatous valve disease. J Am Coll Cardiol 1999;34:2096 –104. Duebener LF, Wendler O, Nikoloudakis N, Georg T, Fries R, Schafers H-J. Mitral-valve repair without annuloplasty rings: results after repair of anterior leaflet versus posterior-leaflet defects using
[22]
[23]
[24]
[25]
559
polytetrafluoroethylene sutures for chordal replacement. Eur J Cardiothorac Surg 2000;17:206–12. Nielsen SL, Timek TA, Lai DT, Daughters GT, Liang D, Hasenkam JM, Ingels NB, Miller DC. Edge-to-edge mitral repair: tension on the approximating suture and leaflet deformation during acute ischemic mitral regurgitation in the ovine heart. Circulation 2001;104(12 Suppl. 1):I29–I35. Gillinov AM, Cosgrove DM, Lytle BW, Taylor PC, Stewart RW, McCarthy PM, Smedira NG, Muehrcke DD, Apperson-Ansen C, Loop FD. Reoperation for failure of mitral valve repair. J Thorac Cardiovasc Surg 1997;113:467–75. Chauvaud SM, Deleuze P, Perier PM, Carpentier AF. Failures in reconstructive mitral valve surgery. Circulation 1986;74(Pt 2): I97–I105. Arts T, Meerbaum S, Reneman R, Corday E. Stresses in the closed mitral valve: a model study. J Biomech 1983;16(7):539–47.
