- Split View
-
Views
-
Cite
Cite
Alexander Bogachev-Prokophiev, Alexander Afanasyev, Sergey Zheleznev, Michael Fomenko, Ravil Sharifulin, Eugenie Kretov, Alexander Karaskov, Mitral valve repair or replacement in hypertrophic obstructive cardiomyopathy: a prospective randomized study, Interactive CardioVascular and Thoracic Surgery, Volume 25, Issue 3, September 2017, Pages 356–362, https://doi.org/10.1093/icvts/ivx152
- Share Icon Share
Abstract
The optimal surgical strategy for concomitant mitral valve intervention during myectomy remains controversial. The purpose of this study was to compare the results of mitral valve replacement versus repair in patients with hypertrophic obstructive cardiomyopathy and severe mitral regurgitation.
Between 2010 and 2013, a total of 88 patients with hypertrophic obstructive cardiomyopathy and severe mitral regurgitation were randomly assigned to undergo either mitral valve replacement or repair in addition to extended myectomy.
Three patients from the repair group were switched to mitral valve replacement after repair failure. There was 1 early death (2.4%) in the replacement group. The resting left ventricular outflow tract gradient was reduced from 89.1 ± 20.4 to 18.3 ± 5.7 mmHg (P < 0.001) and from 96.6 ± 28.1 to 14.7 ± 5.9 mmHg (P < 0.001) in the replacement and repair groups, respectively; there was no significant difference between the groups (P = 0.458). At 2-year follow-up, overall survival was 87.2 ± 4.9% and 96.7 ± 3.3% (P = 0.034); freedom from sudden cardiac death was 95.6 ± 3.1% and 96.7 ± 3.3% (P = 0.615); and freedom from thromboembolic events was 91.2 ± 4.2% and 100%, respectively (P = 0.026).
Both mitral valve repair and valve replacement in addition to extended myectomy are effective methods of surgical treatment in patients with hypertrophic obstructive cardiomyopathy who have severe mitral regurgitation. The benefits of mitral valve repair are better overall survival and a lower rate of thromboembolic events.
ClinicalTrials.gov: NCT02054221.
INTRODUCTION
Surgical septal myectomy is a standard option for patients with hypertrophic obstructive cardiomyopathy (HOCM) who are scheduled for septal reduction therapy. The Morrow procedure is suitable in most cases of HOCM, where there is no mitral valve (MV) pathology. However, in patients who have abnormalities of the papillary muscles and/or secondary chordae, and especially in those with fibrotic anterior leaflet changes, isolated myectomy may be not sufficient to relieve left ventricular outflow tract (LVOT) obstruction and adequately eliminate mitral regurgitation (MR) [1, 2].
According to Yu et al. [3], concomitant MV intervention is required in 11–20% of HOCM patients. Complex MV repair in addition to myectomy may improve the LVOT gradient; however, MV replacement is a simpler surgical alternative. Moreover, the earliest studies showed that MV replacement, either alone or in combination with myectomy, may provide similar results to those of myectomy alone [4, 5].
According to the current guidelines [1], there is no evidence to indicate which is the preferable type of adjunctive procedure in patients with HOCM who undergo concomitant MV surgery. The purpose of our randomized study was to assess MV repair and replacement during extended myectomy in patients with HOCM and severe MR.
PATIENTS AND METHODS
Between November 2010 and August 2013, a total of 198 consecutive HOCM patients with LVOT obstruction underwent surgical myectomy in our centre. In the present study, we report the results from 88 HOCM patients with severe MR who were assigned to undergo either MV repair or replacement in addition to septal myectomy on the day before surgery using a computerized randomization algorithm (ClinicalTrials.gov identifier: NCT02054221). Each of these patients had an LVOT gradient ≥50 mmHg (mean 89.9 ± 27.2 mmHg) and systolic anterior motion (SAM) of the MV at rest.
Eligible participants were adults aged ≥18 years with HOCM who met the indications for operation according to the guidelines of the European Society of Cardiology [1].
Inclusion criteria
Marked septum thickness ≥15 mm measured by echocardiography and/or cardiac magnetic resonance imaging;
Instantaneous peak Doppler LVOT pressure gradient ≥50 mmHg at rest;
Abnormalities of the MV apatus, such as papillary muscle hypertrophy and displacement, fibrotic and retracted secondary chordae, degenerative lesions, etc., revealed by echo and cardiac magnetic resonance imaging;
Resting SAM; and
Severe MR.
The inclusion criterion for participating surgeons was experience of at least 30 septal procedures per year (2 surgeons).
The MV repair group initially included 44 patients: repair was completed in 41, while in 3 cases MV replacement was performed after repair failure. The MV replacement group initially included 44 patients, to which were added the 3 patients with MV repair failure. Baseline data are shown in Table 1. There were no between-group differences in the preoperative characteristics.
. | MV replacement, n = 47 . | MV repair, n = 41 . | P-value . |
---|---|---|---|
Age, years | 50.8 ± 14.3 | 48.3 ± 14.2 | 0.161 |
Male, n (%) | 20 (42.6) | 13 (31.7) | 0.294 |
BSA, m2 | 1.82 ± 0.2 | 1.78 ± 0.3 | 0.822 |
BMI, kg/m2 | 29.3 ± 5.9 | 30.5 ± 5.8 | 0.341 |
Syncope, n (%) | 14 (29.8) | 12 (29.3) | 0.958 |
NYHA Class II, n (%) | 12 (25.5) | 9 (22.0) | 0.694 |
NYHA Class III, n (%) | 33 (70.2) | 29 (70.7) | 0.976 |
NYHA Class IV, n (%) | 1 (2.1) | 3 (7.3) | 0.244 |
Beta-blockers, n (%) | 14 (29.7) | 15 (36.6) | 0.499 |
Verapamil, n (%) | 3 (6.4) | 2 (4.9) | 0.761 |
Disopyramide, n (%) | 4 (8.5) | 2 (4.9) | 0.500 |
Thiazide diuretics, n (%) | 12 (25.5) | 15 (36.6) | 0.262 |
Resting LVOT gradient, mmHg | 90.2 ± 21.1 | 95.3 ± 27.8 | 0.325 |
Septum thickness, mm | 25.5 ± 4.3 | 26.8 ± 4.3 | 0.129 |
Moderate renal impairment, n (%) | 5 (10.6) | 2 (4.9) | 0.319 |
Hypertension, n (%) | 19 (40.4) | 21 (51.2) | 0.310 |
Atrial fibrillation, n (%) | 6 (12.8) | 5 (12.2) | 0.965 |
Five-year risk of SCD, % | 5.4 ± 0.7 | 5.2 ± 0.8 | 0.242 |
Previous ASA therapy, n (%) | 8 (17.0) | 6 (14.6) | 0.760 |
EuroSCORE II, % | 1.8 ± 0.4 | 1.7 ± 0.3 | 0.488 |
. | MV replacement, n = 47 . | MV repair, n = 41 . | P-value . |
---|---|---|---|
Age, years | 50.8 ± 14.3 | 48.3 ± 14.2 | 0.161 |
Male, n (%) | 20 (42.6) | 13 (31.7) | 0.294 |
BSA, m2 | 1.82 ± 0.2 | 1.78 ± 0.3 | 0.822 |
BMI, kg/m2 | 29.3 ± 5.9 | 30.5 ± 5.8 | 0.341 |
Syncope, n (%) | 14 (29.8) | 12 (29.3) | 0.958 |
NYHA Class II, n (%) | 12 (25.5) | 9 (22.0) | 0.694 |
NYHA Class III, n (%) | 33 (70.2) | 29 (70.7) | 0.976 |
NYHA Class IV, n (%) | 1 (2.1) | 3 (7.3) | 0.244 |
Beta-blockers, n (%) | 14 (29.7) | 15 (36.6) | 0.499 |
Verapamil, n (%) | 3 (6.4) | 2 (4.9) | 0.761 |
Disopyramide, n (%) | 4 (8.5) | 2 (4.9) | 0.500 |
Thiazide diuretics, n (%) | 12 (25.5) | 15 (36.6) | 0.262 |
Resting LVOT gradient, mmHg | 90.2 ± 21.1 | 95.3 ± 27.8 | 0.325 |
Septum thickness, mm | 25.5 ± 4.3 | 26.8 ± 4.3 | 0.129 |
Moderate renal impairment, n (%) | 5 (10.6) | 2 (4.9) | 0.319 |
Hypertension, n (%) | 19 (40.4) | 21 (51.2) | 0.310 |
Atrial fibrillation, n (%) | 6 (12.8) | 5 (12.2) | 0.965 |
Five-year risk of SCD, % | 5.4 ± 0.7 | 5.2 ± 0.8 | 0.242 |
Previous ASA therapy, n (%) | 8 (17.0) | 6 (14.6) | 0.760 |
EuroSCORE II, % | 1.8 ± 0.4 | 1.7 ± 0.3 | 0.488 |
MV: mitral valve; BSA: body surface area; BMI: body mass index; NYHA: New York Heart Association; LVOT: left ventricular outflow tract; SCD: sudden cardiac death; ASA: alcohol septal ablation.
. | MV replacement, n = 47 . | MV repair, n = 41 . | P-value . |
---|---|---|---|
Age, years | 50.8 ± 14.3 | 48.3 ± 14.2 | 0.161 |
Male, n (%) | 20 (42.6) | 13 (31.7) | 0.294 |
BSA, m2 | 1.82 ± 0.2 | 1.78 ± 0.3 | 0.822 |
BMI, kg/m2 | 29.3 ± 5.9 | 30.5 ± 5.8 | 0.341 |
Syncope, n (%) | 14 (29.8) | 12 (29.3) | 0.958 |
NYHA Class II, n (%) | 12 (25.5) | 9 (22.0) | 0.694 |
NYHA Class III, n (%) | 33 (70.2) | 29 (70.7) | 0.976 |
NYHA Class IV, n (%) | 1 (2.1) | 3 (7.3) | 0.244 |
Beta-blockers, n (%) | 14 (29.7) | 15 (36.6) | 0.499 |
Verapamil, n (%) | 3 (6.4) | 2 (4.9) | 0.761 |
Disopyramide, n (%) | 4 (8.5) | 2 (4.9) | 0.500 |
Thiazide diuretics, n (%) | 12 (25.5) | 15 (36.6) | 0.262 |
Resting LVOT gradient, mmHg | 90.2 ± 21.1 | 95.3 ± 27.8 | 0.325 |
Septum thickness, mm | 25.5 ± 4.3 | 26.8 ± 4.3 | 0.129 |
Moderate renal impairment, n (%) | 5 (10.6) | 2 (4.9) | 0.319 |
Hypertension, n (%) | 19 (40.4) | 21 (51.2) | 0.310 |
Atrial fibrillation, n (%) | 6 (12.8) | 5 (12.2) | 0.965 |
Five-year risk of SCD, % | 5.4 ± 0.7 | 5.2 ± 0.8 | 0.242 |
Previous ASA therapy, n (%) | 8 (17.0) | 6 (14.6) | 0.760 |
EuroSCORE II, % | 1.8 ± 0.4 | 1.7 ± 0.3 | 0.488 |
. | MV replacement, n = 47 . | MV repair, n = 41 . | P-value . |
---|---|---|---|
Age, years | 50.8 ± 14.3 | 48.3 ± 14.2 | 0.161 |
Male, n (%) | 20 (42.6) | 13 (31.7) | 0.294 |
BSA, m2 | 1.82 ± 0.2 | 1.78 ± 0.3 | 0.822 |
BMI, kg/m2 | 29.3 ± 5.9 | 30.5 ± 5.8 | 0.341 |
Syncope, n (%) | 14 (29.8) | 12 (29.3) | 0.958 |
NYHA Class II, n (%) | 12 (25.5) | 9 (22.0) | 0.694 |
NYHA Class III, n (%) | 33 (70.2) | 29 (70.7) | 0.976 |
NYHA Class IV, n (%) | 1 (2.1) | 3 (7.3) | 0.244 |
Beta-blockers, n (%) | 14 (29.7) | 15 (36.6) | 0.499 |
Verapamil, n (%) | 3 (6.4) | 2 (4.9) | 0.761 |
Disopyramide, n (%) | 4 (8.5) | 2 (4.9) | 0.500 |
Thiazide diuretics, n (%) | 12 (25.5) | 15 (36.6) | 0.262 |
Resting LVOT gradient, mmHg | 90.2 ± 21.1 | 95.3 ± 27.8 | 0.325 |
Septum thickness, mm | 25.5 ± 4.3 | 26.8 ± 4.3 | 0.129 |
Moderate renal impairment, n (%) | 5 (10.6) | 2 (4.9) | 0.319 |
Hypertension, n (%) | 19 (40.4) | 21 (51.2) | 0.310 |
Atrial fibrillation, n (%) | 6 (12.8) | 5 (12.2) | 0.965 |
Five-year risk of SCD, % | 5.4 ± 0.7 | 5.2 ± 0.8 | 0.242 |
Previous ASA therapy, n (%) | 8 (17.0) | 6 (14.6) | 0.760 |
EuroSCORE II, % | 1.8 ± 0.4 | 1.7 ± 0.3 | 0.488 |
MV: mitral valve; BSA: body surface area; BMI: body mass index; NYHA: New York Heart Association; LVOT: left ventricular outflow tract; SCD: sudden cardiac death; ASA: alcohol septal ablation.
The Local Ethics Committee approved the study design, and all the patients provided informed consent. The study was conducted in compliance with the Declaration of Helsinki. The CONSORT flow diagram is shown in Fig. 1.
The primary end point was freedom from MV dysfunction (severe MR recurrence, any prosthesis dysfunction) at 3 years after surgery. The severity of MR was evaluated and defined in accordance with the recommendations of the European Association of Echocardiography [6]. Prosthesis dysfunction was evaluated and defined in accordance with guidelines [7]. Secondary end points were survival, freedom from thromboembolic (TE) events, freedom from sudden cardiac death (SCD) [8] and residual LVOT gradient.
Surgical procedure
Real-time transoesophageal echocardiography (TOE; Phillips iE33, Philips Ultrasound Inc., Reedsville, PA, USA) was performed after the induction of anaesthesia for MV lesion estimation and modelling of an adequate length and depth of resection into the LVOT. The aorta was cross-clamped, and cold crystalloid cardioplegic solution (Custodiol® HTK Solution, Dr Franz Köhler Chemie, Alsbach-Hahnlein, Germany) was used for myocardial protection with antegrade root flow. A transverse aortotomy approach for extended septal myectomy, as described by Messmer [9], was used in all cases. Subsequently, in the repair group, we performed transaortic subvalvular apparatus interventions, including retracted secondary chordae cutting and abnormal papillary muscle release and/or resection. In the replacement group, we preserved the posterior leaflet and implanted On-X prostheses (On-X Life Technologies, Inc., Stafford, TX, USA) in the intra-annular position, using U-stitches with pledgets in anatomic orientation with a 45° rotation about the left ventricular long axis. Control TOE was performed after withdrawal of bypass for routine assessment of LVOT haemodynamics. Direct transaortic catheterization was used for the measurement of pressure gradients. Cardiopulmonary bypass (CPB) was re-established if there was residual moderate-to-severe MR, or if a ventricular septal defect was observed.
Patient follow-up and postoperative management
All patients underwent a transthoracic echocardiographic evaluation before discharge. In total, 87 patients were discharged and followed up periodically by cardiologists and surgeons. After discharge, examinations were scheduled annually. When annual clinic visits were unavailable, follow-up was performed by contact with the referring cardiologist, the patients or their families. Echocardiograms obtained from outside physicians were re-analysed at our institution by the most experienced echocardiographers.
Low-dose aspirin was prescribed postoperatively in the repair group for patients who were in sinus rhythm, as documented by 24-h Holter monitoring. Patients who received a mechanical MV were kept on lifelong anticoagulation with an international normalized ratio target in the range of 2.5–3.5.
Statistical analysis
All data were collected prospectively. Categorical data were expressed as proportions and continuous variables as mean ± standard deviation. Proportions were compared using the χ2 test. If the expected frequency was <5, Fisher’s exact test was applied. Between-group comparisons of continuous variables were performed using an independent-samples t-test or the Mann–Whitney U-test. Estimates of survival, freedom from TE events and SCD were calculated using the Kaplan–Meier method and are reported with 95% confidence intervals (CIs). Curves were compared using the log-rank test. Cox regression models were used to determine univariable and multivariable predictors of overall mortality. The inclusion criterion for the multivariable model was P ≤ 0.2) in the univariable analysis, and the limit for stepwise backward elimination was P < 0.1. SCD and TE events were analysed using competing risk proportional hazards model. Values of P < 0.05 were considered statistically significant. Stata/MP for Windows v. 13.0 (StataCorp. 2013; Stata Statistical Software: Release 13; StataCorp LP, College Station, TX, USA) was used for the statistical analysis.
RESULTS
Intraoperative results
Mean CPB time was 119 (100; 151) and 77 (67; 92) min in the MV replacement and MV repair groups, respectively (P < 0.001). Mean cross-clamp time was significantly longer in the MV replacement group: 73 (56; 94) min versus 44 (36; 58) min in the MV repair group (P < 0.001). In 3 patients who were initially allocated to the MV repair group, a second bypass was established because of repair failure (persistent SAM, moderate MR) and significant peak residual LVOT gradient (>30 mmHg). These patients underwent MV replacement and were reallocated to the respective group. MV morphology is presented in Table 2.
Abnormalities . | Criteria . | MV replacement, n (%) . | MV repair, n (%) . | P-value . |
---|---|---|---|---|
Papillary muscle | Direct to leaflet attachment | 6 (12.7) | 3 (7.3) | 0.456 |
Additional | 8 (17.1) | 4 (9.7) | 0.311 | |
Hypertrophied | 40 (85.1) | 26 (63.4) | 0.022 | |
Chordae | Restrictive chordae of anterior leaflet | 41 (87.2) | 32 (78.1) | 0.240 |
Restrictive chordae of posterior leaflet | 4 (8.5) | 5 (12.2) | 0.456 | |
Leaflet | Thickening | 41 (87.2) | 26 (63.4) | 0.019 |
Elongation | 2 (4.2) | 6 (14.6) | 0.136 |
Abnormalities . | Criteria . | MV replacement, n (%) . | MV repair, n (%) . | P-value . |
---|---|---|---|---|
Papillary muscle | Direct to leaflet attachment | 6 (12.7) | 3 (7.3) | 0.456 |
Additional | 8 (17.1) | 4 (9.7) | 0.311 | |
Hypertrophied | 40 (85.1) | 26 (63.4) | 0.022 | |
Chordae | Restrictive chordae of anterior leaflet | 41 (87.2) | 32 (78.1) | 0.240 |
Restrictive chordae of posterior leaflet | 4 (8.5) | 5 (12.2) | 0.456 | |
Leaflet | Thickening | 41 (87.2) | 26 (63.4) | 0.019 |
Elongation | 2 (4.2) | 6 (14.6) | 0.136 |
MV: mitral valve.
Abnormalities . | Criteria . | MV replacement, n (%) . | MV repair, n (%) . | P-value . |
---|---|---|---|---|
Papillary muscle | Direct to leaflet attachment | 6 (12.7) | 3 (7.3) | 0.456 |
Additional | 8 (17.1) | 4 (9.7) | 0.311 | |
Hypertrophied | 40 (85.1) | 26 (63.4) | 0.022 | |
Chordae | Restrictive chordae of anterior leaflet | 41 (87.2) | 32 (78.1) | 0.240 |
Restrictive chordae of posterior leaflet | 4 (8.5) | 5 (12.2) | 0.456 | |
Leaflet | Thickening | 41 (87.2) | 26 (63.4) | 0.019 |
Elongation | 2 (4.2) | 6 (14.6) | 0.136 |
Abnormalities . | Criteria . | MV replacement, n (%) . | MV repair, n (%) . | P-value . |
---|---|---|---|---|
Papillary muscle | Direct to leaflet attachment | 6 (12.7) | 3 (7.3) | 0.456 |
Additional | 8 (17.1) | 4 (9.7) | 0.311 | |
Hypertrophied | 40 (85.1) | 26 (63.4) | 0.022 | |
Chordae | Restrictive chordae of anterior leaflet | 41 (87.2) | 32 (78.1) | 0.240 |
Restrictive chordae of posterior leaflet | 4 (8.5) | 5 (12.2) | 0.456 | |
Leaflet | Thickening | 41 (87.2) | 26 (63.4) | 0.019 |
Elongation | 2 (4.2) | 6 (14.6) | 0.136 |
MV: mitral valve.
Concomitant procedures included maze IV for atrial fibrillation in 5 (10.6%) patients in the MV replacement and 6 (14.6%) in the MV repair group (P = 0.499). Coronary artery bypass grafting was performed in 3 (6.3%) and 2 (4.9%) patients, respectively (P = 0.644).
Control TOE revealed that the peak LVOT pressure gradient after bypass weaning was 15.6 ± 5.5 mmHg in the MV replacement group and 14.1 ± 5.7 mmHg in the MV repair group (P = 0.252). Direct measurement of LV–aortic gradients did not reveal any significant between-group differences: 9.2 ± 4.6 and 7.4 ± 3.2 mmHg, respectively (P = 0.068). SAM syndrome had resolved in all patients in the MV repair group. The weights of resected myocardium were 6.1 ± 3.2 and 6.5 ± 3.6 g, respectively (P = 0.628).
There was 1 ventricular septal defect after surgery in the MV repair group. Therefore, CPB was re-established and the defect was repaired successfully with a xenopericardial patch. Bleeding due to LV free wall rupture complicated 1 patient from the MV replacement group on the first day after surgery. The patient was successfully treated in the intensive care unit using emergency CPB and LV wall repair. Neither of these complications represented a significant difference between the groups (P = 0.315 and P = 0.282, respectively).
Early postoperative period
There was 1 early death (2.4%) on the 20th day after surgery in the MV replacement group, due to a valve-related TE event (stroke). The patient had stable sinus rhythm and achieved a referenced international normalized ratio of 2.5–3.5.
Complete heart block after extended myectomy occurred in 3 (6.4%) patients in the MV replacement and 2 (4.9%) in the MV repair group (P = 0.282). Permanent dual-chamber pacemaker implantation was performed in all these patients before discharge.
In the MV replacement group, the resting LVOT gradient at discharge had decreased from 89.1 ± 20.4 to 18.3 ± 5.7 mmHg (P < 0.001). In the MV repair group, similar results were observed: the LVOT gradient had reduced from 96.6 ± 28.1 to 14.7 ± 5.9 mmHg (P < 0.001). There was no significant difference between the groups in residual LVOT gradient (P = 0.458). There was no prosthesis dysfunction. No patient had moderate or severe MR at discharge.
Follow-up
Clinical follow-up was completed in all patients (100%). At the last follow-up, New York Heart Association functional class showed a significant decrease from the preoperative value in both groups, with no patients in Class III or IV and no significant difference between groups.
Eight late deaths occurred, 1 in the MV repair group and 7 in the MV replacement group. All patients who died underwent a pathoanatomical examination and autopsy protocols were obtained. One SCD occurred in a patient from the MV repair group 20 months post-myectomy. Three deaths in the MV replacement group were caused by severe pulmonary oedema due to prosthesis thrombosis, while 2 patients had fatal TE complications and SCD occurred in 2 cases. The Kaplan–Meier survival rates at the 2-year follow-up were 87.2 ± 4.9% [95% confidence interval (CI) 73.7–94.0%] and 96.7 ± 3.3% (95% CI 78.6–99.5%) in the MV replacement and MV repair groups, respectively (Fig. 2) (log-rank, P = 0.034). However, freedom from SCD at 24 months was 95.6 ± 3.1% (95% CI 83.4–98.9%) and 96.7 ± 3.3% (95% CI 78.6–99.5%), respectively, and did not differ between the groups (log-rank, P = 0.615; Fig. 3). Competing risk regression analyses regarding risk of SCD confirmed the same result [SHR 0.578 (95% CI, 0.05-6.52), Gray’s test, P = 0.658]. The recalculated risk of SCD at 1-year follow-up was 2.9 ± 0.8 and 2.7 ± 0.9, respectively (P = 0.144). Table 3 shows the univariate analysis of patients’ preoperative characteristics associated with overall mortality. In the multivariable analysis, only NYHA Class II or greater (hazard ratio 4.2, 95% CI 1.5–5.6, P = 0.011) and MV replacement (hazard ratio 2.4, 95% CI 1.2–6.6, P = 0.021) were predictors of late death.
. | Hazard ratio (95% confidence interval) . | P-value . |
---|---|---|
Older age | 1.03 (1.01–2.34) | 0.047 |
Male vs female | 1.12 (0.02–4.88) | 0.388 |
NYHA Class II or greater | 3.21 (1.24–8.15) | 0.024 |
Resting LVOT gradient | 1.02 (0.09–2.43) | 0.128 |
Septum thickness | 1.01 (0.43–9.82) | 0.395 |
Hypertension | 0.45 (0.02–12.21) | 0.448 |
Atrial fibrillation | 2.92 (0.45–4.42) | 0.098 |
Previous alcohol septal ablation therapy | 0.54 (0.09–10.29) | 0.221 |
MV replacement vs repair | 2.7 (1.9–5.5) | 0.002 |
Papillary muscle abnormalities | 0.42 (0.21–12.43) | 0.781 |
Chordal abnormalities | 0.23 (0.01–10.98) | 0.365 |
Leaflet abnormalities | 1.54 (0.25–8.54) | 0.622 |
. | Hazard ratio (95% confidence interval) . | P-value . |
---|---|---|
Older age | 1.03 (1.01–2.34) | 0.047 |
Male vs female | 1.12 (0.02–4.88) | 0.388 |
NYHA Class II or greater | 3.21 (1.24–8.15) | 0.024 |
Resting LVOT gradient | 1.02 (0.09–2.43) | 0.128 |
Septum thickness | 1.01 (0.43–9.82) | 0.395 |
Hypertension | 0.45 (0.02–12.21) | 0.448 |
Atrial fibrillation | 2.92 (0.45–4.42) | 0.098 |
Previous alcohol septal ablation therapy | 0.54 (0.09–10.29) | 0.221 |
MV replacement vs repair | 2.7 (1.9–5.5) | 0.002 |
Papillary muscle abnormalities | 0.42 (0.21–12.43) | 0.781 |
Chordal abnormalities | 0.23 (0.01–10.98) | 0.365 |
Leaflet abnormalities | 1.54 (0.25–8.54) | 0.622 |
NYHA: New York Heart Association; LVOT: left ventricular outflow tract; MV: mitral valve.
. | Hazard ratio (95% confidence interval) . | P-value . |
---|---|---|
Older age | 1.03 (1.01–2.34) | 0.047 |
Male vs female | 1.12 (0.02–4.88) | 0.388 |
NYHA Class II or greater | 3.21 (1.24–8.15) | 0.024 |
Resting LVOT gradient | 1.02 (0.09–2.43) | 0.128 |
Septum thickness | 1.01 (0.43–9.82) | 0.395 |
Hypertension | 0.45 (0.02–12.21) | 0.448 |
Atrial fibrillation | 2.92 (0.45–4.42) | 0.098 |
Previous alcohol septal ablation therapy | 0.54 (0.09–10.29) | 0.221 |
MV replacement vs repair | 2.7 (1.9–5.5) | 0.002 |
Papillary muscle abnormalities | 0.42 (0.21–12.43) | 0.781 |
Chordal abnormalities | 0.23 (0.01–10.98) | 0.365 |
Leaflet abnormalities | 1.54 (0.25–8.54) | 0.622 |
. | Hazard ratio (95% confidence interval) . | P-value . |
---|---|---|
Older age | 1.03 (1.01–2.34) | 0.047 |
Male vs female | 1.12 (0.02–4.88) | 0.388 |
NYHA Class II or greater | 3.21 (1.24–8.15) | 0.024 |
Resting LVOT gradient | 1.02 (0.09–2.43) | 0.128 |
Septum thickness | 1.01 (0.43–9.82) | 0.395 |
Hypertension | 0.45 (0.02–12.21) | 0.448 |
Atrial fibrillation | 2.92 (0.45–4.42) | 0.098 |
Previous alcohol septal ablation therapy | 0.54 (0.09–10.29) | 0.221 |
MV replacement vs repair | 2.7 (1.9–5.5) | 0.002 |
Papillary muscle abnormalities | 0.42 (0.21–12.43) | 0.781 |
Chordal abnormalities | 0.23 (0.01–10.98) | 0.365 |
Leaflet abnormalities | 1.54 (0.25–8.54) | 0.622 |
NYHA: New York Heart Association; LVOT: left ventricular outflow tract; MV: mitral valve.
There were 5 TE events in the MV replacement group: 3 patients had valve thrombosis and 2 suffered ischaemic stroke. Figure 4 indicates that the rate of freedom from TE events was significantly higher in the MV repair group (100%) than in the MV replacement group (91.2 ± 4.2%, 95% CI 78.3–96.6%; log-rank, P = 0.026). The cumulative incidence function (SHR and P-value (Gray’s test) for TE events was estimated due to separation of the data.
No paravalvular leakage was observed during follow-up in the MV replacement group. In the MV repair group, 2 (5%) surviving patients had moderate MR recurrences at last follow-up, and another had mild or trace MR. No reoperation was performed during follow-up. Thus, we did not find any significant between-group difference in the primary end point, freedom from MV dysfunction.
At the last echocardiographic examination, the provoked LVOT gradient was 17.5 ± 3.8 mmHg in the MV replacement group and 18.3 ± 4.2 mmHg in the MV repair group (P = 0.913).
DISCUSSION
Currently, transaortic septal myectomy is the gold standard for the surgical treatment of HOCM patients with severe drug refractory symptoms [1]; however, MV replacement remains an alternative approach [4, 5]. Our study is the first prospective randomized trial to compare MV repair and MV replacement in HOCM patients with severe MR. In our study, mean CPB cross-clamp times were significantly longer in the MV replacement group; this can be explained by the need to open and close the left atrium and the additional time required for the replacement procedure, whereas MV management in the repair group was performed through a transaortic approach.
According to the Mayo Clinic’s experience, MV intervention concomitant with septal myectomy is a rare necessary procedure [10]. A retrospective analysis showed that only 174 of 1993 patients really required MV management; however, it should be noted that only 57.5% of these patients had MR Grade ≥3 (of 4) preoperatively. The authors commented on the difficulties in assessing the true preoperative MR mechanism in HOCM patients, distinguishing between common SAM and other lesions involving the MV apparatus. They reported that approximately 43% of patients (75 of 174 who underwent MV surgery) had an MV lesion that was not discovered until intraoperative checking. The high rate of concomitant MV intervention in our study (88 of 198 patients) can be explained by the careful elective enrolment. We recruited only patients with HOCM who had severe MR, and we used preoperative TOE and cardiac magnetic resonance screening in all patients to confirm that the MR was not caused by SAM alone. Only patients with marked MV abnormalities were eligible for randomization. In patients who are scheduled for myectomy, with basal septum thickening and mild or moderate MR, in the absence of significant MV pathology, we usually avoid any adjunctive procedures, as adequate extended myectomy effectively abolishes SAM and restores a satisfactory LVOT gradient in most patients.
We suppose that different varieties of concomitant MV pathology in the HOCM subset might require different MV adjunctive procedures, including leaflet plication [11, 12], leaflet extension [13–15], retention [16, 17], subvalvular management [2] or simple MV replacement [4, 5]. It is impossible to perform a single repair technique that will effectively eliminate SAM and MR, while restoring an adequate LVOT gradient, for all kinds of MV abnormality. However, MV replacement is a universal, simple and reproducible surgical alternative.
The superiority of MV repair is well established [18]. There is strong evidence of a survival benefit in patients undergoing MV repair for degenerative MV disease [19, 20]. However, in severe ischaemic MR, MV replacement yields better 2-year results than MV repair [21]. The optimal surgical strategy for concomitant MR during septal reduction therapy in HOCM patients is still controversial. Vassileva et al. [22] examined the strategy for MV surgery in the HOCM subset using the largest all-payer database in the USA. Among 1255 HOCM patients who underwent concomitant MV surgery, the repair rate was only 17.2%. Moreover, they found a significant difference in hospital mortality: 0.00% for the repair group and 11.8% for patients who required replacement (P < 0.05). The largest retrospective clinical study [10] demonstrated better late survival for repair (80.0%) vs replacement (55.2%) at 10-year follow up (P = 0.002). However, the authors provided no explanation of this superiority in survival. In our study, all patients who died underwent autopsy. The superiority of survival in the MV repair group in our study can be explained by better freedom from valve-related TE events. We did not find a difference in the freedom from SCD, which might have occurred as a result of cardiomyopathy-related arrhythmias. Both the recalculated and the observed risks of SCD were low in both groups, with no between-group differences. In the present study, in addition to SAM, one of the most common causes of MR was fibrotic changes of the anterior leaflet due to chronic mitral–septal contact; this was especially marked in aged patients and in those who had undergone previous alcohol septal ablation. Such MV lesions are characterized by a deficiency in anterior leaflet area; thus, leaflet extension by pericardial patch could be the method of choice for MV repair. This technique was found to be superior to myectomy alone in a study by Kofflard et al. [14], in terms of postoperative residual MR, residual SAM and functional class improvement. Acceptable mid-term results were previously described by van der Lee et al. [13]; however, their study did not include a between-group comparison. A long-term benefit of anterior mitral leaflet extension in the HOCM subset with adequate MV function and survival comparable to that of the general population were reported by Vriesendorp et al. [15]. We consider that extension of the anterior leaflet is a complex technique that entails an additional risk of late repair failure due to pericardial patch deterioration or dehiscence. This is possibly one of the reasons for the wide use of MV replacement in HOCM patients, with the expectation of good results. In the MV repair group, we extensively used the secondary chordae resection technique described by Ferrazzi et al. [2]. We believe that the cutting of thickened secondary chordae attached to the middle part of the anterior leaflet allows extension of the anterior leaflet’s surface. The transaortic secondary chordal cutting procedure is a simple and reproducible manoeuvre.
The high rate of TE events in our study is related to the low compliance with anticoagulation therapy; only 62% of patients had achieved the target international normalized ratio at their 1-year follow-up visit. We do not believe that the observed between-group differences were related to the On-X mitral valves. Furthermore, the TE event rate for MV replacement in this specific patient group is considerably higher than the general TE event rate for MV replacement in our practice. It should be noted that MV replacement with a high-profile biological prosthesis is not a suitable strategy, given the additional risk of LVOT obstruction [23, 24]. Therefore, MV replacement with a mechanical valve is the compelled strategy.
Limitations
This was a single-centre study and was limited by the relatively small sample size. However, our results are consistent with those of other investigators. Furthermore, the follow-up period was only 2 years. A longer follow-up might reveal more differences between the clinical outcomes of the 2 techniques. Further studies with larger patient populations and longer follow-up will be needed to confirm our findings.
CONCLUSION
Both MV repair and MV replacement effectively eliminate SAM and LVOT obstruction during extended septal myectomy and improve the functional capacity in patients with HOCM and MV pathology. The benefits of MV repair are better overall survival and a lower rate of TE events; however, no advantages were observed in terms of SCD, functional capacity or LVOT gradient relief. Our data suggest that MV repair should be performed in most patients who are scheduled for myectomy and who require concomitant MV management for severe MR that is not caused by SAM alone.
ACKNOWLEDGEMENTS
We would like to thank Editage for assistance with the editing of the manuscript.
Funding
This work was supported by a grant from the President of the Russian Federation [MD-5046.2015.7].
Conflict of interest: none declared.
REFERENCES
Author notes
Presented at the 29th Annual Techno-College Meeting of the European Association for Cardio-Thoracic Surgery, Barcelona, Spain, 1 October 2016.