Author + information
- Received September 14, 2014
- Revision received October 27, 2014
- Accepted November 4, 2014
- Published online February 10, 2015.
- Philippe B. Bertrand, MD, MSc∗,†,
- Frederik H. Verbrugge, MD∗,†,
- David Verhaert, MD∗,
- Christophe J.P. Smeets, MSc†,
- Lars Grieten, PhD, MSc∗,†,
- Wilfried Mullens, MD, PhD∗,†,
- Herbert Gutermann, MD‡,
- Robert A. Dion, MD, PhD‡,
- Robert A. Levine, MD§ and
- Pieter M. Vandervoort, MD∗,†∗ ()
- ∗Department of Cardiology, Ziekenhuis Oost-Limburg, Genk, Belgium
- †Faculty of Medicine and Life Sciences, Hasselt University, Diepenbeek, Belgium
- ‡Department of Cardiac Surgery, Ziekenhuis Oost-Limburg, Genk, Belgium
- §Cardiac Ultrasound Laboratory, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. Pieter M. Vandervoort, Department of Cardiology, Ziekenhuis Oost-Limburg, Schiepse Bos 6, Genk 3600, Belgium.
Background Restrictive mitral valve annuloplasty (RMA) for secondary mitral regurgitation might cause functional mitral stenosis, yet its clinical impact and underlying pathophysiological mechanisms remain debated.
Objectives The purpose of our study was to assess the hemodynamic and clinical impact of effective orifice area (EOA) after RMA and its relationship with diastolic anterior leaflet (AL) tethering at rest and during exercise.
Methods Consecutive RMA patients (n = 39) underwent a symptom-limited supine bicycle exercise test with Doppler echocardiography and respiratory gas analysis. EOA, transmitral flow rate, mean transmitral gradient, and systolic pulmonary arterial pressure were assessed at different stages of exercise. AL opening angles were measured at rest and peak exercise. Mortality and heart failure readmission data were collected for at least 20 months after surgery.
Results EOA and AL opening angle were 1.5 ± 0.4 cm2 and 68 ± 10°, respectively, at rest (r = 0.4; p = 0.014). EOA increased significantly to 2.0 ± 0.5 cm2 at peak exercise (p < 0.001), showing an improved correlation with AL opening angle (r = 0.6; p < 0.001). Indexed EOA (EOAi) at peak exercise was an independent predictor of exercise capacity (maximal oxygen uptake, p = 0.004) and was independently associated with freedom from all-cause mortality or hospital admission for heart failure (p = 0.034). Patients with exercise EOAi <0.9 cm2/m2 (n = 14) compared with ≥0.9 cm2/m2 (n = 25) had a significantly worse outcome (p = 0.048). In multivariate analysis, AL opening angle at peak exercise (p = 0.037) was the strongest predictor of exercise EOAi.
Conclusions In RMA patients, EOA increases during exercise despite fixed annular size. Diastolic AL tethering plays a key role in this dynamic process, with increasing AL opening during exercise being associated with higher exercise EOA. EOAi at peak exercise is a strong and independent predictor of exercise capacity and is associated with clinical outcome. Our findings stress the importance of maximizing AL opening by targeting the subvalvular apparatus in future repair algorithms for secondary mitral regurgitation.
Secondary mitral regurgitation (MR) in ischemic and/or dilated heart disease is associated with an unfavorable prognosis (1,2). Restrictive mitral valve annuloplasty (RMA) has evolved as the gold standard treatment for severe secondary MR (3–6); however, recurrence of MR after RMA has an incidence of approximately 30% (7), and mitral valve replacement (MVR) has been proposed as a potential alternative in selected patients (8–11). In addition, several studies have demonstrated the occurrence of moderate mitral stenosis (defined as mean transmitral pressure gradient [TMG] >5 mm Hg or mitral valve area <1.5 cm2) after RMA (12–16). Functional mitral stenosis after RMA is generally attributed to undersizing of the annular ring, yet the subvalvular apparatus might play a role as well. Importantly, the impact of such functional mitral stenosis on exercise capacity and clinical outcome remains unclear, mainly because most studies have based stenosis grading on the mean TMG at rest (13,14,16–19). However, this parameter is certainly flow dependent, potentially masking severe stenosis in low-flow patients (15). There are currently few data on the use of less flow-dependent parameters such as the effective orifice area (EOA) to quantify stenosis in RMA patients.
The aim of this study was to investigate the hemodynamic, functional, and clinical impact of EOA after RMA and its relationship with diastolic anterior leaflet (AL) tethering at rest and during exercise.
Study design and study population
We included consecutive patients who underwent RMA with a complete semirigid Physio ring (Edwards Lifesciences, Irvine, California), undersized by 1 or 2 sizes to obtain a minimal coaptation length of 8 mm in the A2-P2 segment, for secondary MR (Carpentier class IIIb, i.e., systolic leaflet restriction) between July 2007 and September 2012 at a single tertiary care center (Ziekenhuis Oost-Limburg, Genk, Belgium). Exclusion criteria were structural leaflet abnormalities at surgical inspection, inability to undergo a supine bicycle exercise test, more than mild aortic regurgitation (AR) (vena contracta width >3 mm), and more than mild MR recurrence (vena contracta width >3 mm) at rest or during exercise. Eligible patients underwent a comprehensive resting transthoracic echocardiography (TTE) examination, followed by a semisupine symptom-limited bicycle exercise test with concomitantly performed TTE. The study complied with the Declaration of Helsinki, the protocol was approved by the local ethics committee, and written informed consent was obtained from all participating patients.
Resting and exercise echocardiography was performed with a commercially available system (Philips Medical Systems, IE33, Andover, Massachusetts). Standard 2-dimensional and Doppler images were acquired in the left lateral decubitus position and stored digitally for offline analysis in the CardioView software (TomTec Imaging Systems, Unterschleissheim, Germany). All measurements were averaged over 3 consecutive cardiac cycles for patients in sinus rhythm and 5 consecutive cycles for patients with atrial fibrillation according to the guidelines of the American Society of Echocardiography. The modified Simpson’s biplane method was used to calculate ejection fraction at rest and peak exercise (20). Peak and mean TMG were calculated using the modified Bernoulli equation on the continuous-wave transmitral Doppler signal, with EOA calculated by the continuity equation (21). Systolic pulmonary artery pressure (sPAP) was calculated using the modified Bernoulli equation on the transtricuspid continuous-wave signal, while adding an estimate of right atrial pressure (22). Because patients with more than mild MR and AR were excluded, mean transmitral flow rate was defined as the ratio of left ventricular (LV) stroke volume (measured from pulsed wave Doppler in the LV outflow tract) over diastolic filling time. During each stage of exercise, LV stroke volume, transmitral flow rate, peak and mean TMG, sPAP, and EOA were assessed. The mitral AL opening angle was measured on 2-dimensional TTE in an apical long-axis view, both at rest and at peak exercise. Measures were taken at peak E-wave, and the angle of the maximal excursion of the AL was measured with respect to the plane of the prosthetic annular ring (Central Illustration).
All participating patients underwent a symptom-limited graded bicycle test in the semisupine position on a tilting exercise table. Workload was initiated at 20 W, with increments of 20 W every 3 min. In patients with poor general condition, an adjusted protocol was applied with 10 W of initial workload and increments of 10 W every 3 min. Blood pressure, a 12-lead electrocardiogram, ergospirometry (Jaeger, Würzburg, Germany), and echocardiography measurements were recorded at each stage.
All-cause mortality and heart failure admissions (defined as hospitalization because of signs or symptoms of congestion that warranted treatment with parenteral drugs) were registered in all study patients from the day of surgery until July 31, 2014, which yielded a follow-up of at least 20 months after surgery for every patient. This follow-up included retrospective clinical data collected during the postoperative period before the study visit, as well as prospective data from the study visit until July 31, 2014.
Results of continuous variables were expressed as mean ± SD if normally distributed or otherwise by median and interquartile range. Normality was assessed by the Shapiro-Wilk statistic. The paired or unpaired Student t test and Wilcoxon signed rank test were used whenever appropriate. Categorical variables were expressed as percents and compared with Fisher exact test. Linear regression models were used to assess the correlation between TMG, EOA, and the square of transmitral flow rate. Predictors of maximal oxygen uptake (VO2max) and EOA with a p value <0.1 at univariate analysis were entered in multiple linear regression models. Cox proportional hazards regression was used to assess variables associated with freedom from all-cause mortality or heart failure readmission since surgery, and variables with p < 0.1 were entered in a multivariate Cox regression model. An assumption was made that hemodynamic data at the time of study were representative for the entire follow-up period. Cumulative survival rates were calculated according to the Kaplan–Meier method, and groups were compared with the log-rank test. Receiver operating characteristic curves were used to determine area under the curve, sensitivity and specificity of different parameters, and cutoffs for the prediction of impaired exercise capacity (VO2max <15 ml/kg/min). Statistical significance was always set at a 2-tailed probability of p < 0.05. All statistical analyses were performed with the Statistical Package for Social Sciences version 20.0 (SPSS Inc., Chicago, Illinois).
Of 103 screened patients, 27 patients had died. Nine were excluded because of structural leaflet abnormalities at surgical inspection, and 24 patients did not perform an exercise test for various reasons (orthopedic or neurological limitations, n = 9; distance to hospital, n = 3; and refusal to participate, n = 12). Four patients were excluded from the analyses because they had recurrent MR or AR at rest or during exercise at the time of the study visit. Accordingly, the final study population consisted of 39 patients. Table 1 summarizes their baseline characteristics. None of the study patients had angina pectoris at rest or during the exercise test, there were no new ischemic alterations on the 12-lead electrocardiogram monitoring during exercise, and no obvious new wall motion abnormalities were observed at peak exercise in standard apical views.
Transmitral pressure-flow relationship at rest and during exercise
Echocardiographic measures at rest and at maximal exercise for all 39 patients after mean follow-up of 33 ± 17 months are presented in Table 2. Mean and peak TMG, cardiac output, and sPAP increased significantly during exercise. The evolution of the mean TMG with respect to cardiac output at each stage of exercise is displayed in Figure 1A. There was a strong and significant correlation between the mean TMG and the square of the cardiac output (r = 0.81; p < 0.001), as well as the square of the transmitral flow rate (r = 0.81; p < 0.001) (Figure 1B). This quadratic relationship is in agreement with fundamental hydraulic notions derived from the Bernoulli equation and the principle of continuity, in which the pressure gradient across a stenotic orifice EOA is known to be a function of the flow rate F squared (23,24), as displayed in Equations 1 and 2:(Equation 1)(Equation 2)
A display of this theoretical hydraulic relationship for various EOA orifices ranging from 0.5 to 3.5 cm2 is superposed in Figure 1B.
Relationship between transmitral flow rate and EOA
The mean EOA at rest was 1.5 ± 0.4 cm2 (EOA indexed for body surface area, or EOAi, 0.8 ± 0.2 cm2/m2) in RMA patients. On the basis of this resting value and assuming a constant EOA, the expected mean TMG at maximal flow rate would be much higher than the observed mean gradient at maximal flow rate. During exercise, however, an increase in EOA was observed with increasing flow rate (quadratic correlation: r = 0.78; p < 0.001) (Figure 2). EOA at maximal exercise was 2.0 ± 0.5 cm2 (EOAi 1.0 ± 0.3 cm2/m2, p < 0.001 for the change from rest).
Diastolic AL opening angle
The AL opening angle at rest (68° ± 10°) showed a moderate correlation with EOA at rest (r = 0.4; p = 0.014). However, exercise EOA was much better correlated to AL opening angle at peak exercise (r = 0.6; p < 0.001). Larger increases in AL opening angle during exercise were associated with higher relative increases in EOA (r = 0.42; p = 0.001). Univariate linear regression identified annuloplasty ring size (r = 0.4; p = 0.007), AL opening angle at peak exercise (r = 0.6; p < 0.001), and peak cardiac output (r = 0.5; p = 0.002) as determinants of EOAi at peak exercise. In the multivariate model, AL opening angle at peak exercise (p = 0.037) remained the strongest determinant of EOAi, independent of ring size (p = 0.087) and peak cardiac output (p = 0.137).
Impact of EOAi on exercise capacity and clinical outcome
VO2max was 15.2 ± 4.3 ml· kg−1 ·min−1 in the population overall. In multivariate analysis, patient age (p = 0.004) and exercise EOAi (p = 0.004) emerged as independent predictors of VO2max. At a cutoff value of 0.9 cm2/m2, exercise EOAi (area under the curve 0.750) had 86% sensitivity and 62% specificity to predict an exercise capacity lower than 15 ml/kg/min. During the entire follow-up period (55 ± 16 months), 11 patients were admitted to the hospital for decompensated heart failure, and 2 patients died. In a multivariate Cox regression model to assess freedom from all-cause mortality or heart failure admission (Table 3) exercise EOAi (p = 0.034) and age (p = 0.008) were independently associated with outcome. Patients with exercise EOAi below the above-mentioned cutoff of 0.9 cm2/m2 (n = 14) had significantly worse outcome than patients with higher exercise EOAi (n = 25), as shown in the Central Illustration.
This study quantified the transmitral pressure-flow relationship in patients with secondary MR treated with RMA and demonstrated an increase in EOA during exercise. Exercise EOAi was independently associated with outcome and was a stronger predictor of exercise capacity than was EOAi at rest, which stresses the importance of exercise echocardiography in this patient population. In addition, our findings challenge the concept that functional mitral stenosis from RMA for secondary MR solely results from a small annular size. If that were the case, the stenotic valve area would be fixed and less responsive to exercise. Instead, our findings suggest that the observed stenosis after RMA has an important subvalvular component as well (in addition to ring size), dictated by the diastolic restriction of anterior leaflet opening, likely caused by diastolic leaflet tethering.
Diastolic leaflet tethering: Pathophysiological insights
Secondary MR results from LV motion abnormalities or dilation, causing an imbalance between closing forces and papillary muscle tethering, thereby interfering with normal systolic coaptation (25). These “culprit” LV abnormalities have diastolic implications as well, not only impairing LV relaxation but also inhibiting diastolic leaflet opening caused by severe leaflet tethering. This phenomenon can be aggravated by insertion of a prosthetic ring, anteriorly displacing the posterior annulus and increasing the importance of the AL during diastole (which then acts like the lid of a pedal-operated waste bin). AL tethering caused by papillary muscle displacement decreases the opening angle of this lid and obstructs inflow (Figure 3). During exercise, increasing flow and left atrial pressure might (in part) overcome the tethering force and increase the opening angle to some extent. Importantly, however, the AL “opening reserve” might also be determined by the LV response to exercise (17). In patients with a positive LV response (i.e., a significant decrease in LV end-systolic volume), papillary muscle tethering decreases and AL opening and EOA increase (Central Illustration).
Importantly, increasing the AL opening angle will not only increase the geometric orifice area at the tip of the restricted leaflets but also has an impact on the coefficient of flow contraction. An alteration of the inflow angulation during exercise, caused by a conformational change in the subvalvular apparatus, is known to increase the coefficient of contraction (26). Thus, any exercise-induced increase in leaflet angulation will result in both an increase in geometric orifice area and an increase in the coefficient of contraction, which compounds the effect on EOA.
Comparison with previous work
In a similar RMA population, Magne et al. (12) also observed a significant increase in EOA during exercise with a concomitant rise in LV ejection fraction (Table 4). Kubota et al. (17), on the other hand, observed a significant decrease in EOA during exercise after RMA. However, LV contractile reserve was severely impaired in this population, with the ejection fraction decreasing during exercise. Moreover, the study by Kubota et al. (17) suggested diastolic subvalvular tethering as an important contributor to functional mitral stenosis after RMA. Finally, Fino et al. (27) found no exercise-induced increase in EOA in RMA patients despite an increase in ejection fraction. Increasing ejection fraction during exercise, however, might be characterized by increased LV end-diastolic volume rather than a decrease in end-systolic volume. This might have caused a difference in exercise tethering in the patient group in the study by Fino et al. (27) compared with the RMA population in the study by Magne et al. (12) and the present study. Moreover, in the patient group in the study by Fino et al. (27), exercise EOAi was an independent predictor of exercise sPAP (as a surrogate of exercise capacity).
In MVR patients, severe valve prosthesis–patient mismatch (defined as EOAi <0.9 cm2/m2) is known to be associated with postoperative pulmonary hypertension and worse outcomes (28,29). Our findings indicate that similar mismatch hemodynamics and physiology can also be observed in patients after RMA, with 2 important differences, however. First, EOAi was an independent predictor of outcome in RMA patients not at rest but at peak exercise. We hypothesize that in MVR patients, the EOA is fixed during exercise and therefore is an identical predictor of outcome both at rest and peak exercise. Second, because the mechanism of inflow obstruction is related to diastolic leaflet restriction, patients with adverse remodeling and LV dilation will have the lowest EOAi at peak exercise, further aggravating exercise capacity and outcome in a vicious circle. This might explain the independently strong association between exercise EOAi, exercise capacity, and outcome in this small sample population.
The demonstration of a dynamic mitral valve area in RMA patients because of diastolic leaflet restriction may add important insights with respect to functional capacity, exercise hemodynamics, and treatment options in this patient population. First, the increase in TMG during exercise is more attenuated in RMA patients than in patients with similar resting hemodynamics but with a more fixed orifice, as seen in organic mitral stenosis (30,31) and mechanical MVR (32). When describing and grading the degree of functional stenosis after RMA on the basis of resting data alone, it is important to realize the possible dynamic behavior of the EOA, especially in patients with beneficial LV remodeling. In case of doubt, exercise echocardiography with assessment of EOA at peak exercise should be applied. Second, our findings strongly support further research of concomitant subvalvular procedures for patients with secondary MR to relieve leaflet tethering. Prior studies have demonstrated that ongoing leaflet tethering is involved in the mechanism of MR recurrence (33,34). Our study shows leaflet tethering probably plays a role in diastolic inflow restriction as well. Relieving tethering might therefore increase repair durability, reduce the degree of undersizing needed to obtain the target coaptation length, and reduce the functional stenosis after surgery. Importantly, the increase in EOA tended to reach a plateau (2.0 to 2.5 cm2) at higher flow rates. We hypothesize this plateau is influenced by the effective orifice of the ring in the absence of leaflet tethering. Indeed, in multivariate analysis, ring size remained associated with exercise EOA, independent of leaflet angle. Therefore, all efforts to ensure a durable repair while relieving tethering and increasing annuloplasty ring size should be subject to further research.
This was a single-center study with limited sample size and a small endpoint count (i.e., there was a risk of overfitting the Cox regression model). Therefore, our results should be considered hypothesis generating. Clinical follow-up for outcome analysis started immediately after surgery and thus preceded the exercise study visit. We assumed the hemodynamic and functional measures at the study visit were representative for the entire postoperative follow-up period. This assumption is on the basis of prior RMA studies demonstrating stable hemodynamics (mean transmitral gradient at rest) from early after surgery until 18-month follow-up (3), as well as evidence of early postoperative improvement in New York Heart Association functional class that remained stable for up to 7 years of follow-up in patients undergoing coronary artery bypass grafting with mitral valve annuloplasty for secondary MR (35). Finally, patients with recurrent MR after RMA at rest or during exercise were excluded from analysis to ensure the validity of the continuity equation. Because the mechanism of MR recurrence is commonly related to ongoing leaflet tethering (33) and restricted leaflet motion (34), one could hypothesize that the change in EOA and AL opening angle with increasing flow would be more limited or even negative in these excluded patients.
This study demonstrated that in patients with secondary MR after RMA, the EOA increases with increasing transmitral flow rate despite fixed annular size. Diastolic AL tethering plays a key role in this dynamic process, with increasing AL opening during exercise being associated with higher exercise EOA. EOAi at peak exercise is a strong and independent predictor of exercise capacity and is independently associated with outcome. Our findings stress the importance of maximizing AL mobility by targeting the subvalvular apparatus in future repair algorithms for secondary MR.
COMPETENCY IN MEDICAL KNOWLEDGE: Ongoing leaflet tethering after mitral valve repair in patients with secondary MR contributes both to recurrent MR and to diastolic inflow obstruction post-operatively.
TRANSLATIONAL OUTLOOK: Further studies are needed to determine whether relief of leaflet tethering in patients undergoing repair of secondary MR improves the durability of repair, increases diastolic orifice area, and is associated with better long-term clinical outcomes.
The authors thank the technicians and nursing staff of the cardiac ultrasound laboratory (R. Reyskens, K. Cuyvers, K. Machiels, and L. Jacobs) for their excellent support in the data collection.
Dr. Bertrand is supported by a grant of the Research Foundation–Flanders (FWO, 11N7214N). Dr. Bertrand, Dr. Verbrugge, Mr. Smeets, Dr. Grieten, Dr. Mullens, and Dr. Vandervoort are researchers for the Limburg Clinical Research Program UHasselt-ZOL-Jessa, supported by the foundation Limburg Sterk Merk, Hasselt University, Ziekenhuis Oost-Limburg, and Jessa Hospital. Dr. Dion has received consulting fees from Edwards Lifesciences, Johnson & Johnson, Sorin Biomedica, Medtronic, and St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- anterior (mitral) leaflet
- aortic regurgitation
- effective orifice area
- indexed effective orifice area
- left ventricle
- mitral regurgitation
- mitral valve replacement
- restrictive mitral valve annuloplasty
- transmitral pressure gradient
- transthoracic echocardiography
- maximal oxygen uptake
- Received September 14, 2014.
- Revision received October 27, 2014.
- Accepted November 4, 2014.
- American College of Cardiology Foundation
- Lamas G.A.,
- Mitchell G.F.,
- Flaker G.C.,
- et al.,
- for the Survival and Ventricular Enlargement Investigators
- Grigioni F.,
- Enriquez-Sarano M.,
- Zehr K.J.,
- Bailey K.R.,
- Tajik A.J.
- Vahanian A.,
- Alfieri O.,
- Andreotti F.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.
- Magne J.,
- Girerd N.,
- Sénéchal M.,
- et al.
- Kouris N.,
- Ikonomidis I.,
- Kontogianni D.,
- Smith P.,
- Nihoyannopoulos P.
- Magne J.,
- Sénéchal M.,
- Mathieu P.,
- Dumesnil J.G.,
- Dagenais F.,
- Pibarot P.
- Kainuma S.,
- Taniguchi K.,
- Daimon T.,
- et al.
- Martín C.E.,
- Castaño M.,
- Gomez-Plana J.,
- Gualis J.,
- Comendador J.M.,
- Iglesias I.
- Lang R.M.,
- Bierig M.,
- Devereux R.B.,
- et al.
- Levine R.A.,
- Schwammenthal E.
- Gilon D.,
- Cape E.G.,
- Handschumacher M.D.,
- et al.
- Li M.,
- Dumesnil J.G.,
- Mathieu P.,
- Pibarot P.
- Magne J.,
- Mathieu P.,
- Dumesnil J.G.,
- et al.
- Voelker W.,
- Berner A.,
- Regele B.,
- et al.
- Leavitt J.I.,
- Coats M.H.,
- Falk R.H.
- Mihaljevic T.,
- Lam B.K.,
- Rajeswaran J.,
- et al.