Access to the PDF text

Free Article !

Archives of cardiovascular diseases
Volume 111, n° 8-9
pages 528-533 (août 2018)
Doi : 10.1016/j.acvd.2017.12.005
Received : 10 May 2017 ;  accepted : 11 December 2017
Clinical research

Systematic transoesophageal echocardiography after mitral valve replacement: Rates and determinants of paravalvular regurgitation
Échographie trans-œsophagienne systématique après remplacement valvulaire mitrale : prévalence et détérminants des fuites périprothétiques

Constance Verdonk a, , Claire Cimadevilla b, Laurent Lepage b, Fanny Perez c, Walid Ghodbane b, Richard Raffoul b, Soleiman Alkhoder b, Patrick Nataf b, Alec Vahanian a, David Messika-Zeitoun a
a Department of Cardiology, Bichat Hospital, AP–HP, 46, rue Henri-Huchard, 75018 Paris, France 
b Department of Cardiac Surgery, Bichat Hospital, AP–HP, 75018 Paris, France 
c Cardiology Master 2, Bichat Hospital, AP–HP, 75018 Paris, France 

Corresponding author.

With the emergence of transcatheter mitral valve replacement, it appears crucial to provide contemporary references with which this new technology can be compared. At our institution, transoesophageal echocardiography is systematically performed before discharge after surgical mitral valve replacement.


To evaluate the rate and determinants of paravalvular regurgitation after surgical mitral valve replacement.


We collected medical history, indication, type of surgery and in-hospital outcome in all consecutive patients who underwent a mitral valve replacement in the past 2 years at our institution. Paravalvular regurgitation was assessed semiquantitatively using transoesophageal echocardiography before discharge.


We enrolled 399 patients (mean age 61±16 years; 58% women; 27% with a history of cardiac surgery). Mitral valve replacement was performed mainly for rheumatic disease (44%). Most patients were severely symptomatic (70% in New York Heart Association class III/IV). A mechanical prosthesis was implanted in 60% and a bioprosthesis in 40%. In-hospital mortality was 10%. Transoesophageal echocardiography was performed in 310 patients (77%); the main reasons for not performing transoesophageal echocardiography were frailty (n =40, 10%), early death (n =19, 5%) and contraindication for transoesophageal echocardiography (n =10, 3%). The overall rate of paravalvular regurgitation was 8% (n =25); a grade ≥2 was observed in five patients (2%), and two patients had to be reoperated on. Mitral annular calcification was the main factor associated with paravalvular regurgitation (P =0.01).


Surgical mitral valve replacement was associated with significant in-hospital mortality and morbidity. Using systematic transoesophageal echocardiography assessment, paravalvular regurgitation was not uncommon (8%), and was significantly linked to mitral annulus calcification. However, clinically significant leakage (≥grade 2) was rare (2%).

The full text of this article is available in PDF format.

Avec le développement du remplacement valvulaire mitral percutané, il devient indispensable de fournir des référentiels pour évaluer les nouvelles techniques. Dans notre hôpital, une échographie trans-œsophagienne est systématiquement réalisée avant la sortie de l’hôpital chez les patients ayant bénéficié d’un remplacement valvulaire mitral chirurgical.


Évaluer la fréquence et les déterminants des fuites para prothétiques après une chirurgie de remplacement valvulaire mitral.


Nous avons collecté les antécédents, l’indication et le type de chirurgie, les résultats de la chirurgie chez tous les patients ayant bénéficié d’un remplacement valvulaire mitral au cours des deux dernières années dans notre hôpital. Les fuites paravalvulaire ont été évaluées semi quantitativement par le biais de l’échographie trans-œsophagienne réalisée avant la sortie.


Nous avons inclus 399 patients (âge moyen 61±16 ans ; 58 % de femmes ; 27 % ayant un antécédent de chirurgie cardiaque). La pathologie mitrale principale était le rhumatisme articulaire aigu (44 %). La majorité des patients étaient très symptomatiques (70 % des patients stade II/IV de la NYHA). Une prothèse mécanique a été implantée dans 60 % des cas contre 40 % pour les prothèses biologiques. La mortalité intra-hospitalière était de 10 %. L’échographie trans-œsophagienne a pu être pratiquée chez 310 patients (77 %), les raisons principales de ne pas faire l’examen étaient la fragilité (n =40, 10 %), le décès précoce (n =19, 5 %) et la présence de contre-indication (n =10, 3 %). Le taux de fuite paraprothétique était de 8 % (n =25) ; un grade ≥2 était présent chez 5 patients (2 %), deux patients ont dû être réopérés. Les calcifications de l’anneau mitral étaient le principal facteur associé à la survenue d’une fuite para prothétique (p =0,01).


Le remplacement valvulaire mitral chirurgical est associé à une morbi-mortalité intra-hospitalière significative. Par le biais d’une évaluation échographique trans-œsophagienne, les fuites paraprothétiques ne sont pas exceptionnelles (8 %) et leur survenue est liée dans notre étude à la présence de calcification de l’anneau mitral. Cependant les fuites cliniquement parlantes (≥grade 2) restent rares (2 %).

The full text of this article is available in PDF format.

Keywords : Paravalvular regurgitation, Echocardiography, Mitral valve replacement

Mots clés : Fuite paravalvulaire, Échocardiographie, Remplacement valvulaire mitral

Abbreviations : MAC, MVR, PVR, TMVR, TOE, TTE


Surgical mitral valve replacement (MVR) is challenged by emerging techniques for patients considered contraindicated to or at high risk from surgery, such as those who have had previous cardiac surgery or mitral annular calcification [1]. Paravalvular regurgitation (PVR) after mitral valve replacement needs to be evaluated specifically in these subgroups, as they will probably be the target population for transcatheter mitral valve replacement (TMVR). Post-procedural PVR, already known in transaortic valve replacement, will indubitably arise as an important issue in TMVR [2]. Previous work has established a rate of postoperative PVR of 12% using transoesophageal echocardiography (TOE), but in this study only a few patients underwent MVR [3]. Other series concerning MVR showed a PVR rate of 13%, but the technique used to define PVR was not specified [4]. In our institution, TOE is performed systematically after MVR, before discharge, as part of the clinical workup.

The aim of the present study was to evaluate the rate and determinants of PVR after surgical MVR, assessed by systematic TOE evaluation, thus providing a contemporary reference that the new technology can be compared with.

Study population

All consecutive patients who underwent isolated or combined surgical MVR over the past 2 years in Bichat Hospital (Paris, France) were included. Patients undergoing mitral valve repair were excluded. Medical history was collected retrospectively. The aetiology of mitral disease was classified as rheumatic disease, infective endocarditis, prosthesis dysfunction or other (functional regurgitation, iatrogenic valve disease, congenital valve disease, tumoural valve disease, complication of transaortic valve replacement). The type of prosthesis implanted, associated surgery and duration of intervention were recorded. Surgery was considered urgent when performed within 48hours of hospital admission. Postoperative prolonged inotropic support was defined by the need for intravenous inotropic support for >24hours after the end of surgery. In-hospital mortality and postoperative complications, such as tamponade, postoperative prolonged inotropic support (>24hours), supraventricular arrhythmia and the need for pacemaker implantation, were also recorded.


Comprehensive preoperative transthoracic echocardiography (TTE) provided information about the preoperative ejection fraction and the presence of mitral annular calcification. The ejection fraction was evaluated visually or according to the Simpson method. Mitral annular calcification was visually assessed qualitatively in the short-axis parasternal view. Systolic pulmonary artery pressure was measured by using the maximal gradient of tricuspid regurgitation, to which right atrium pressure (range 5–20mmHg) was added.

PVR was diagnosed based on postoperative systematic TOE, performed by experienced operators before discharge, and was semiquantitatively graded from mild to severe using an integrative approach, as recommended [5]. Indeed, to quantify regurgitation we used standardized analysis of the mitral valve regarding the vena contracta of the jet in TOE, the extension of the jet and (when feasible) the area of the regurgitant orifice. TOE was performed on patients who were awake or slightly sedated (using midazolam). iE33 (Philips, Amsterdam, the Netherlands) and Vivid 7 (GE Healthcare, Chicago, IL, USA) two-dimensional probes were used for postoperative TOE. To evaluate the localization of the regurgitation we analysed the valve carefully in two-dimensional segments according to Carpentier.

Statistical analysis

Mean values were calculated for all continuous variables. Numbers of patients and percentages were calculated for categorical variables. Continuous variables were compared using the t -test for independent samples. Determinants of PVR were assessed using logistic regression. A P -value<0.05 was considered statistically significant. Statistical analysis was performed using JMP 7 software (SAS Institute, Cary, NC, USA). The authors are solely responsible for the design and conduct of this study, all analyses and its final content.

Study population

During the 2-year study period, 399 MVRs (mean age 61±16 years; 42% men) were performed at our institution (Table 1). History of supraventricular arrhythmia was common (n =177, 44%), and cardiac surgery had been performed in 107 patients (27%) (mitral valve surgery in 90 patients [23%]). Most patients were severely symptomatic, and 282 patients were in New York Heart Association class III/IV (70%). The main aetiology of mitral valve disease was rheumatic disease (n =174, 44%), and the main surgical indication was mitral regurgitation (n =182, 46%). A mechanical prosthesis was implanted in most patients (n =238, 60%). An isolated MVR was performed in a minority of patients (n =96, 24%); a tricuspid annuloplasty was also performed in 215 patients (54%), an aortic valve replacement was also performed in 114 patients (28%) and a coronary artery bypass graft was also performed in 37 patients (9%). Surgery was considered urgent in 109 patients (27%), and was mainly secondary to active endocarditis (n =47, 43%).

In-hospital outcome

The in-hospital mortality rate was 10% (n =39). Isolated mitral surgery was associated with an in-hospital mortality rate of 5%, whereas combined surgery led to an in-hospital mortality rate of 11% (P =0.07). Mortality was caused equally by sepsis (n =11) and cardiogenic shock (n =11) (56%), rupture of the mitral annulus (n =4) and myocardial rupture (n =4) (20%) and mesenteric ischaemia (n =5, 12%). One patient died as a result of haemorrhagic shock secondary to psoas haematoma, one patient died as a result of cerebral anoxic lesion and two patients presented unexpected death.

The most frequent complication was the occurrence of supraventricular arrhythmia (n =105, 26%). Pericardial effusion was also frequently found on systematic postoperative echocardiography, and led to drainage in a fifth of the study population (n =71, 18%). Prolonged inotropic support was necessary in 184 patients (46%). Conduction disorders were present in 117 patients (29%); 25 of these patients (6%) required implantation of a pacemaker.

Rates and determinants of PVR

Postoperative TTE analysis was not used to diagnose PVR because of its lack of sensitivity. TOE was performed in 310 patients (77%). We chose not to collect results from peroperative TOE, as PVR can be over-rated or underestimated depending on the patient's blood pressure and loading condition during surgery. The main reasons for not performing TOE were frailty (n =40/89, 45%), early death (n =19/89, 21%) and contraindication to TOE, such as neurological impairment or history of oesophageal or orotracheal tract lesion (n =10/89, 11%); other patients either refused the examination or were transferred to another institution before TOE could be performed (Figure 1). As expected, patients who did not undergo postoperative TOE were significantly older (68±16 vs. 60±16 years; P =0.001) and had a higher in-hospital mortality rate (24% vs. 6%; P <0.0001). It is important to note that none of these patients had TTE evidence of PVR.

Figure 1

Figure 1. 

Flowchart of study population according to transoesophageal echocardiography and paravalvular regurgitation.


Among the 310 patients who underwent TOE, the overall PVR rate was 8% (n =25), including five patients (2%) who presented moderate-to-severe regurgitation. The PVR rate was relatively stable among various subsets. For patients with no history of cardiac surgery, the PVR rate was slightly lower than for patients with a history of cardiac surgery, although the difference was not significant (11% vs. 7%; P =0.15); similar results were found for patients with a history of mitral valve surgery (11% vs. 7%; P =0.23) (Figure 2). The PVR rate was not significantly different when MVR was performed for infective endocarditis compared with for non-infectious reasons (3% vs. 9%; P =0.12). PVR was also not significantly more frequent in isolated MVR compared with MVR combined with another procedure (12% vs. 7%; P =0.13). Urgent surgery was not associated with a higher incidence of PVR than non-urgent surgery (both 8%; P =0.77). Preoperative mitral annular calcification (MAC) was the main factor associated with PVR (18% vs. 6%; P =0.01) (Table 1). In multivariable analysis (logistic regression), MAC was the only determinant of PVR (P =0.03).

Figure 2

Figure 2. 

Rates of paravalvular regurgitation in various subgroups. A. History of cardiac surgery. B. Mitral annulus calcification (MAC). C. Endocarditis-related mitral valve replacement (MVR). D. Combined surgery.


Two patients had to be reoperated on because of severe PVR. One had undergone four previous surgeries and had a fragile mitral annulus and early postoperative periprosthetic leakage; the heart team's consideration of this case led to a heart transplant. The other patient was reoperated on 1 month after discharge because of persistent symptoms; the surgical findings were a tear in a calcified mitral annulus. PVR percutaneous closure was not done in any of the patients because of the size of the periprosthetic defect and the procedural risk.


In a contemporary series of patients undergoing MVR, the main aetiology was rheumatic disease, history of heart surgery was common and the in-hospital mortality rate was relatively high. Using systematic TOE during the immediate postoperative period, the PVR rate was 8% (mostly mild to moderate). The main determinant of postoperative PVR was the presence of MAC, whereas a history of cardiac surgery or surgery for acute infective endocarditis were not determinants. Our cohort of patients presented few cases of endocarditis, which may have led to an underestimation of the link between PVR and endocarditis, which has been described previously [4].

There are few studies on contemporary surgical MVR, and historical series are often cited for comparison, with a wide range of PVR rates (ranging from 1% to 60%) [2, 3, 6]. In addition, the PVR rate is often reported at follow-up, and limited data exist on PVR rates immediately after surgery. In a large consecutive and contemporary series of patients who underwent MVR, the rate of PVR was 8% overall, and was relatively stable among all subsets tested [3]. It is worth noting that in our study, TOE, which provides more accurate assessment of MVR complications than TTE, was performed systematically and was not clinically guided. In addition, TOE can accurately assess the location of the regurgitation, and thus can establish its paravalvular location (as opposed to intraprosthetic location). Finally, TOE was performed just before discharge under local anaesthesia, and thus under normal loading conditions. Our study provides important findings in terms of diagnosis, localization of leakage and real immediate postoperative incidence of PVR.

Our work confirms the direct link between MAC and PVR. MAC is a well-known challenge for surgeons, and has been significantly linked to PVR in previous studies [7, 8, 9]. Interestingly, in this subset of patients with MAC, TMVR has recently been raised as a potential alternative to surgery [7, 8, 9]. The present study clearly identifies the surgical challenge caused by MAC, and provides important references with which TMVR can be compared. We were unable to identify other determinants of postoperative PVR, possibly because of the lack of power and sample size.

The present study deserves several comments. Firstly, it was a single-centre study with inherent bias, especially regarding recruitment and patient characteristics. Secondly, TOE was not performed in 23% of our population, mainly because of frailty. Importantly, none of the deaths observed during the in-hospital period was deemed to be related to prosthetic dysfunction. Thirdly, the long-term outcome of PVR and haemolysis were not assessed in the present study, and we only focused on the incidence of PVR during the postoperative period. In the immediate postoperative period, a low level of haemoglobin is hard to interpret, and is poorly correlated to haemolysis. Analyses of lactate dehydrogenase and haptoglobin were not performed systematically during our study. Finally, no long-term outcome was evaluated. It would be interesting to evaluate the clinical impact and echocardiographic evolution of PVR by follow-up of our cohort. In addition, late occurrence of PVR was not assessed. Significant PVR is an important issue, as it may be responsible for an increase mortality and morbidity; reported results of percutaneous closure of PVR seem promising [10, 11].


Surgical MVR was associated with significant in-hospital mortality, mainly as a result of immediate postoperative shock and morbidity. PVR was not uncommon, with an incidence rate of 8%, and was significantly linked to MAC. However, clinically relevant leakage (moderate-to-severe and severe) was rare. The present study provides a reference with which emerging TMVR can be compared.

Sources of funding

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Disclosure of interest

The authors declare that they have no competing interest.


Millan X., Skaf S., Joseph L., and al. Transcatheter reduction of paravalvular leaks: a systematic review and meta-analysis Can J Cardiol 2015 ;  31 : 260-269 [cross-ref]
Duncan B.F., McCarthy P.M., Kruse J., and al. Paravalvular regurgitation after conventional aortic and mitral valve replacement: a benchmark for alternative approaches J Thorac Cardiovasc Surg 2015 ;  150 : 860-868
Ionescu A., Fraser A.G., Butchart E.G. Prevalence and clinical significance of incidental paraprosthetic valvar regurgitation: a prospective study using transoesophageal echocardiography Heart 2003 ;  89 : 1316-1321 [cross-ref]
Genoni M., Franzen D., Vogt P., and al. Paravalvular leakage after mitral valve replacement: improved long-term survival with aggressive surgery? Eur J Cardiothorac Surg 2000 ;  17 : 14-19 [cross-ref]
Nishimura R.A., Otto C.M., Bonow R.O., and al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines J Am Coll Cardiol 2014 ;  63 : e57-e185
Meloni L., Aru G., Abbruzzese P.A., and al. Regurgitant flow of mitral valve prostheses: an intraoperative transesophageal echocardiographic study J Am Soc Echocardiogr 1994 ;  7 : 36-46 [cross-ref]
Fassa A.A., Himbert D., Brochet E., and al. Transseptal transcatheter mitral valve implantation for severely calcified mitral stenosis JACC Cardiovasc Interv 2014 ;  7 : 696-697 [cross-ref]
Guerrero M., Urena M., Pursnani A., and al. Balloon expandable transcatheter heart valves for native mitral valve disease with severe mitral annular calcification J Cardiovasc Surg (Torino) 2016 ;  57 : 401-409
Himbert D., Bouleti C., Iung B., and al. Transcatheter valve replacement in patients with severe mitral valve disease and annular calcification J Am Coll Cardiol 2014 ;  64 : 2557-2558 [cross-ref]
Cho I.J., Hong G.R., Lee S., Byung-Chul C., Ha J.W., Chung N. Predictors of prognosis in patients with mild to moderate paravalvular leakage after mitral valve replacement J Card Surg 2014 ;  29 : 149-154 [inter-ref]
Krishnaswamy A., Kapadia S.R., Tuzcu E.M. Percutaneous paravalvular leak closure-imaging, techniques and outcomes Circ J 2013 ;  77 : 19-27 [cross-ref]

© 2018  Elsevier Masson SAS. All Rights Reserved.
EM-CONSULTE.COM is registrered at the CNIL, déclaration n° 1286925.
As per the Law relating to information storage and personal integrity, you have the right to oppose (art 26 of that law), access (art 34 of that law) and rectify (art 36 of that law) your personal data. You may thus request that your data, should it be inaccurate, incomplete, unclear, outdated, not be used or stored, be corrected, clarified, updated or deleted.
Personal information regarding our website's visitors, including their identity, is confidential.
The owners of this website hereby guarantee to respect the legal confidentiality conditions, applicable in France, and not to disclose this data to third parties.
Article Outline