Article Text

Update on percutaneous mitral commissurotomy
  1. Maria Carmo P Nunes1,
  2. Bruno Ramos Nascimento1,
  3. Lucas Lodi-Junqueira1,
  4. Timothy C Tan2,
  5. Guilherme Rafael Sant'Anna Athayde1,
  6. Judy Hung3
  1. 1Hospital das Clinicas, School of Medicine, Federal University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil
  2. 2Department of Cardiology, Blacktown Hospital, University of Western Sydney, Sydney, New South Wales, Australia
  3. 3Cardiac Ultrasound Lab, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Maria Carmo P Nunes, Hospital das Clinicas, School of Medicine, Federal University of Minas Gerais, Belo Horizonte, Avenida Professor Alfredo Balena, 190, Room 246, Belo Horizonte, MG 30130-100, Brazil; mcarmo{at}


Percutaneous mitral commissurotomy (PMC) is the first-line therapy for managing rheumatic mitral stenosis. Over the past two decades, the indications of the procedure have expanded to include patients with unfavourable valve anatomy as a consequence of epidemiological changes in patient population. The procedure is increasingly being performed in patients with increased age, more deformed valves and associated comorbidities. Echocardiography plays a crucial role in patient selection and to guide a more efficient procedure. The main echocardiographic predictors of immediate results after PMC are mitral valve area, subvalvular thickening and valve calcification, especially at the commissural level. However, procedural success rate is not only dependent on valve anatomy, but a number of other factors including patient characteristics, interventional management strategies and operator expertise. Severe mitral regurgitation continues to be the most common immediate procedural complication with unchanged incidence rates over time. The long-term outcome after PMC is mainly determined by the immediate procedural results. Postprocedural parameters associated with late adverse events include mitral valve area, mitral regurgitation severity, mean gradient and pulmonary artery pressure. Mitral restenosis is an important predictor of event-free survival rates after successful PMC, and repeat procedure can be considered in cases with commissural refusion. PMC can be performed in special situations, which include high-risk patients, during pregnancy and in the presence of left atrial thrombus, especially in centres with specialised expertise. Therefore, procedural decision-making should take into account the several determinant factors of PMC outcomes. This paper provides an overview and update of PMC techniques, complications, immediate and long-term results over time, and assessment of suitability for the procedure.

Statistics from


Rheumatic heart disease remains a public health concern especially in developing countries, where it accounts for over a million premature deaths annually.1 Rheumatic mitral stenosis (RMS) results from inflammatory response to acute rheumatic fever. Inflammation of the mitral valve apparatus gives rise to commissural fusion, leaflet thickening, chordal shortening or a combination of all these.2–4 RMS is a slowly progressive disease, characterised by a prolonged latent phase between the acute rheumatic fever and the development of valve stenosis. Once mild stenosis has developed, further narrowing is typical. The rate of progression of stenosis can vary significantly5 as the natural history of RMS is different across the world.4 The management of RMS is dependent on stage of the disease process. Patients who are symptomatic with a haemodynamically significant MS should be considered for valve intervention.

Percutaneous mitral commissurotomy: temporal trends

Percutaneous mitral commissurotomy (PMC) has had a significant impact on the management of RMS, replacing surgical therapy as the primary choice of intervention. In addition, indications for PMC have greatly expanded due to improved patient selection and technical advances.6–9 Although no randomised trials have been performed to ascertain the best timing of intervention, PMC is generally offered to appropriate patients once symptoms develop and before significant reduction in functional capacity and development of severe pulmonary hypertension. In addition, there is increasing evidence demonstrating the potential benefits of early intervention in selected patients with favourable valve morphology.10 High procedural success rate associated with low procedure-related complications has been reported in patients with ideal valve anatomy. Moreover, early PMC may also decrease the occurrence of embolic events.10 ,11

Over the past two decades, the indications of the procedure have been expanded to include patients with unfavourable valve anatomy as a consequence of changes in epidemiology and advances in invasive techniques.6 ,7 ,12 ,13 The procedure is increasingly being performed in patients with increased age and more deformed valves. The systematic use of the Inoue stepwise technique has simplified the procedural steps and improved its efficacy as the operators became familiar with this technique As a consequence, the results have been maintained over the years despite suboptimal PMC candidates. However, a recent cross-sectional study of PMC in the USA from 1998 to 2010 reported trends of increasing procedural complication rates related to increasing age and associated comorbidities.12 On the other hand, the patients with MS from low-income and middle-income countries are younger than those from Western countries with a higher procedural success rate. Nevertheless, the use of PMC is limited in developing countries for economical reasons and open surgical commissurotomy is still a less-expensive alternative therapy.1

Echocardiographic assessment of suitability for PMC

A detailed evaluation of valve morphology, including leaflet thickness and mobility, degree and localisation of calcifications, and the extent of the subvalvular involvement is essential to determine the appropriate intervention. Specifically, aside from the presence of valve calcification, the extent and the location of calcification is also an important factor to take into account in the procedural decision-making.6 ,14–17 However, the use of echocardiography for assessing valve calcification lacks specificity since dense fibrous nodules and calcification generally have the same aspect.

A number of echocardiographic parameters have been used to assess mitral valve anatomy and function. There are also a number of scoring systems proposed in order to optimise patient selection and to predict outcomes.15–28 Studies published over the past two decades, summarised in online supplementary table S1, analysed different variables to determine suitability for PMC. The most widely currently used method was proposed by Wilkins et al.18 The Wilkins score encompasses an assessment of leaflet mobility, thickening, calcification and subvalvular thickening. A high total score including a semi-quantitative assessment of each parameter predicted an insufficient valve opening, which was associated with low event-free survival rate.20 However, this score fails to predict outcomes in those with scores of 9–11, which represent an intermediate population that have progressively been selected for the procedure.18 Another scoring system commonly used in Europe is based on the extent of subvalvular disease and mitral calcification as assessed by fluoroscopy.19 Nevertheless, these scoring systems have a poor predictive value in determining outcome after PMC, especially for scores in the intermediate range and are based on criteria that can be subjective. Furthermore, no direct comparisons between existing scoring systems have been made.

More recently, a new score for predicting procedural success that used quantitative parameters (leaflet displacement and asymmetry in commissural remodelling) was shown to be useful in predicting outcomes after PMC.14 This new scoring system accurately predicted procedural success both as a composite outcome of increase in valve area and development of worsening mitral regurgitation (MR), when applied to a population from an endemic region. However, all echocardiographic scores have significant limitations and an integrative approach is mandatory.

Three-dimensional (3D) echocardiography has also been increasingly used for measurement of mitral valve area (MVA) in a more reproducible manner than two-dimensional in inexperienced hands, and for assessment of commissural opening29 (figure 1). Commissural calcification, which affects the success of PMC, is difficult to appreciate by 3D echocardiography. Further studies are still needed to verify whether assessment of the mitral valve using 3D echocardiography translates into improved patient selection and outcomes after PMC.

Figure 1

Three-dimensional echocardiography of the mitral valve with severe rheumatic mitral stenosis prior to percutaneous mitral commissurotomy (A). Same patient after percutaneous mitral commissurotomy showing that anterolateral commissure was opened (B).

Since commissural opening is the primary mechanism of increase of the MVA (figure 2), the pattern and localisation of valvular calcification is an important determinant of PMC results. Previous studies have shown that calcification localised to the commissure level were associated with a lower procedural success rate, a higher occurrence of MR and a lower mid-term survival.6 ,16 ,17 ,30 Based on this, the choice for PMC or surgery is frequently debated in patients with commissural calcification. Dreyfus et al6 evaluated the impact of the extension and location of mitral valve calcification on PMC results and found that commissural calcification was associated with a lower procedural success rate. Similarly, Bouleti et al,13 who compared the outcome of 314 patients with calcific MS undergoing PMC with 710 patients with non-calcified valves, showed good immediate results in 80% of the patients with calcific MS. However, the extent of valve calcification was only assessed by fluoroscopy in this study, which unfortunately did not allow differentiation between commissural and leaflet calcification. The conclusions of recent papers are not to preclude PMC only because of calcification even at the commissural level. Clinical characteristics that predict results should be taken into account, especially if a PMC is not performed in these patients, the only alternative would be mitral valve replacement.

Figure 2

Two-dimensional echocardiography at the parasternal short-axis view of the mitral valve prior to percutaneous mitral commissurotomy (PMC) showing fused both commissures (A). Same patient after successful PMC, both commissures were fully split (B). Mitral regurgitation jets originating at both commissures are shown by colour Doppler flow mapping at the apical two-chamber view (C). LV, left ventricle; RV, right ventricle.

There is increasing evidence demonstrating that clinical outcomes following PMC is not only dependent on mitral valve anatomy, but a range of factors, including clinical characteristics, echocardiographic parameters and interventional management strategies.20 Additionally, the rheumatic disease process can also cause additional distinct structural derangements of the mitral valve that may influence outcome of PMC. While favourable outcome of PMC is expected if the main underlying feature is limited to commissural fusion, however, if the subvalvular or leaflets are severely involved, then PMC success is less likely. The pattern of leaflet involvement is also an important determinant of MR after PMC. Padial et al15 showed that heterogeneously thickened leaflets with thick areas coexisting with thin or almost normal zones predicted the risk of MR.

PMC: technique

The most common PMC techniques are the transseptal approaches, using either a double-balloon or the Inoue balloon system. The Inoue balloon technique is the most commonly used technique worldwide because it is faster, less laborious and requires less fluoroscopy time than the double-balloon technique.31 However, there is no significant difference in postprocedural results between the two antegrade techniques.32

An overview of the Inoue balloon technique

Transseptal puncture

All antegrade approaches begin with the crucial step of transseptal catheterisation. Most commonly, transseptal puncture is performed via the right common femoral vein access, but if unavailable, then the left femoral or right jugular veins can also be used as alternative access approach. Although the puncture of the fossa ovalis is generally quite safe, the risk of a transseptal approach lies in the possibility of puncturing an adjacent structure.

One of the simplest and most widely used methods to locate the site for optimal transseptal puncture is the use of fluoroscopic landmarks (figure 3). Alternatively, periprocedural transthoracic, transesophageal or intracardiac echocardiography may help in identifying the location for transseptal puncture.

Figure 3

(A) In the anteroposterior projection, the puncture site should be halfway from the distance of the catheter and the right border of the left atrium and one vertebral body height lower to the aortic valve plane (seen by inserting a pig tail catheter in the non-coronary aortic sinus). (B) In the 40° right anterior oblique projection, confirm if the needle is positioned neither close to the aortic root (too anterior) nor close to the free wall of the right atrium (too posterior). The puncture site should be 1–3 cm below the midpoint of a line connecting the posterior wall of the aorta to the back wall of the right atrium.

Selection of appropriate balloon size

Selection of the appropriate balloon diameter is crucial for the success of the procedure. The formula using patient's height is a simple and effective way to determine the optimal size of balloon to use (balloon diameter [mm]=height [cm]/10+10).33 Based on this value, the operator selects the right catheter size, bearing in mind that a lower volume balloon than what is calculated may be used initially but larger balloons may subsequently be needed if the immediate result is suboptimal. Ideally, the stenosed mitral valve should be initially dilated using balloons 1 mm lower than the calculated diameter in very tall patients since the relationship between the patient's height to the diameter of the mitral valve may not necessarily be linear. This is particularly relevant in patients with critical MS since the risk of undesired injury to the mitral valve during PMC is potentially higher.

The PMC procedure

After crossing the mitral orifice (figure 4), the balloon is positioned within the apical portion of the left ventricle. The distal portion of the balloon is then inflated and the catheter is pulled back until it reaches the mitral valve plane before fully inflating the balloon to dilate the valvar orifice. After each balloon inflation, MVA (planimetry), commissural splitting and the degree of MR should be assessed by transthoracic echocardiography. The operator may make 1 mL increments in the balloon volume with every subsequent dilation until an adequate MVA has been achieved or an increase in MR is seen. This led to the abandonment of cardiac catheterisation to assess pressures and left ventriculography in most cases.

Figure 4

(A) In an anterior oblique projection, slightly pull back the balloon catheter and rotate the J-shaped stylet counterclockwise, helping its tip turning anteriorly towards the mitral valve. (B) When it reaches the orifice, the balloon is moved forward across the mitral orifice and the stylet is slightly pulled back. An alternative technique for crossing the mitral orifice may be useful in very large left atrium or in unusual anatomical features. (C) After inserting the balloon catheter into the left atrium, rotate the stylet clockwise to face the balloon tip towards the posterior inferior atrial wall and (D) pull back the stylet, advancing only the balloon catheter through the mitral orifice.

Immediate results of PMC in the current era

In the past decades, there have been some substantial changes in the pattern of immediate outcomes of PMC, related to the cumulative experience with the technique, patient selection and the different criteria defining procedural success. Although the basic interventional techniques of PMC (Inoue and double-balloon) have not changed dramatically within the past two decades, improved patient selection and echo-guided procedure may account for better outcomes.

The measures of immediate success based on a number of studies differ on three aspects: (1) post-PMC MVA: while many studies use the absolute criterion of MVA ≥1.5 cm2, a 50% increase in the baseline area has also been long accepted as a measure of success; (2) MR: some studies consider the more strict criterion of an increase in angiographic MR as a suboptimal result, while other series use the cut-off of Seller's grade of a ≥2/3 points increase; more recent studies assess the increase in the severity of MR as assessed by echocardiography instead of ventriculography; and (3) other major complications that require emergency surgery.

Studies in the past decade have considered PMC success as an increase in MVA (>50% or MVA ≥1.5 cm2), without a significant increase in MR. The main echocardiographic predictors of immediate procedural results are related to MV anatomy including baseline MVA, subvalvular thickening and valve calcification, especially at the commissural level (table 1). Procedural success rates varied widely among studies, even when similar definitions were applied. In general, these rates ranged from 65% to 80%, with some authors reporting rates >90% in considerable sized cohorts.25 ,34–36 A list of clinical, haemodynamic and morphological variables associated with immediate success is shown in table 1 and online supplementary tables S1 and S2. The most common independent predictors were gender (worse outcomes observed in women20 ,23 ,37), age and New York Heart Association (NYHA) functional class. Some studies have also demonstrated higher event rates in patients with atrial fibrillation (AF)—markedly embolic complications and stroke—although this finding was not unanimous.36 ,38

Table 1

Predictors of immediate results after percutaneous mitral commissurotomy

The acute complication rates of PMC are also relatively low in most of the published series.12 ,39 However, in a 13-year perspective of procedures in the USA, the sum of all cardiac, vascular, neurological, renal and metabolic acute complications reached 15.9%.12 The rate of complications of PMC also depends on the number of cases performed per centre. Severe MR continues to be the most common immediate complication with reported rates between 1.5% and 7.5%, rarely exceeding 10%,8 ,40 which remains unchanged in the past decades.30 There are several different mechanisms for the MR seen after PMC, which include wide valve opening with no recognisable structural abnormalities (figure 5), leaflet tearing (figure 6) and chordae rupture. Clinical tolerance to iatrogenic acute severe MR is variable, with a few patients requiring emergency surgery, usually below 5% in contemporaneous cohorts (see online supplementary table S3), while the majority appears to tolerate the MR well.41 ,42 The pattern of valve thickening and commissural calcification is associated with MR risk.14 ,15 ,30

Figure 5

Two-dimensional echocardiography at the parasternal short-axis view of the mitral valve before (A) and after percutaneous mitral commissurotomy (B and C). After the procedure, an excessive opening of the mitral valve occurred with no recognisable structural abnormalities of the valve (B). Significant mitral regurgitation is shown by colour Doppler flow mapping at the apical four-chamber view (C). LV, left ventricle; RA, right atrium; RV, right ventricle.

Figure 6

Two-dimensional echocardiography at the parasternal short-axis view before (A) and after percutaneous mitral commissurotomy (B). A new echocardiographic dropout (B, arrow) at the anterior mitral leaflet was seen as a result of leaflet tearing after the procedure. Transesophageal echocardiography confirmed the leaflet tearing (C, arrow) with significant mitral regurgitation (D). LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.

Other complications arising from the PMC include cardiac tamponade, embolic events and bleeding.

Long-term outcome after PMC

There is significant interest in the long-term procedural outcomes studies since PMC is now considered the standard treatment modality for suitable patients with RMS.20 The long-term outcome after PMC is mainly determined by the immediate procedural results.

Mitral restenosis has been shown to be an important determinant of late adverse events after successful PMC. Restenosis has been shown to be a time-dependent process with variable incidence rates30 ,43–45 due to the large variability related to the definition used to define restenosis and the follow-up durations. Song et al45 clearly showed the immediate post-PMC MVA as the most important predictor of both restenosis and major clinical events.

Postprocedural severe MR has also been shown to be associated with worse long-term event-free survival. A study in which 876 consecutive patients with MS underwent PMC showed that preprocedural and postprocedural MR grades independently predicted the long-term clinical outcomes after PMC.46 Mild MR originating from the commissures is a marker of complete commissural opening, which is associated with late functional improvement.47

There are presently few longitudinal studies that examine the changes in MVA and MR long term after PMC. One study including 561 patients showed that more than two-thirds of patients remained in good clinical condition and free of any major event 7 years after PMC30 Another more recent study showed that in 912 consecutive patients undergoing PMC followed up for up to 20 years, 30% still had good functional results.48

Several predictors of long-term outcome of PMC have been reported. Table 2 summarises the predictors of long-term outcomes in recent studies published in the past decade, and online supplementary table S3 shows all the relevant studies on long-term outcome after PMC. Factors that have been identified include preprocedural parameters such as (1) patient's age as an independent predictor of both immediate and long-term outcomes,20 ,49 (2) NYHA class IV at presentation, (3) MVA, (4) previous surgical commissurotomy, (5) presence of MR, (6) Wilkins echo score ≤8 and (7) presence of tricuspid regurgitation.50 Postprocedural parameters associated with long-term outcomes include (1) immediate result of the PMC procedure, where the post-PMC MVA of ≥1.8 cm2 has been shown to have good outcomes;45 (2) presence of postprocedural MR; (3) mean mitral gradient after PMC; (4) pulmonary artery pressure;20 (5) development of significant late tricuspid regurgitation even in the absence of pulmonary hypertension;51 (6) progressive increase of left atrial (LA) volume even after successful PMC;52 and (7) the presence of AF.53

Table 2

Predictors of long-term outcome after percutaneous mitral commissurotomy (PMC)

Special indications for PMC

PMC has also been shown to be feasible and successful in patients deemed to likely have unfavourable outcomes from rheumatic disease. In high-risk patients where surgery is contraindicated, it remains as the only definitive therapy available. Critically ill patients who are too unstable to undergo a surgical procedure can be treated with the PMC as definitive therapy or as a bridge to surgery with acceptable results.

RMS is usually associated with aortic and tricuspid valve disease, and the management for these other valve lesions follow the recommendations concerning the predominant valve disease.8 ,39 PMC can be indicated in severe MS associated with moderate aortic stenosis with expectant conduct to the aortic involvement, which is different if surgical intervention is being considered. In the patients with severe functional tricuspid regurgitation, mitral valve surgery and tricuspid valve repair appear to have better clinical outcomes than PMC and the surgical option should be considered.54 It is also established that patients with these two conditions have more severe disease.50 In cases of severe MS and tricuspid regurgitation, especially in patients in sinus rhythm when the LA is not very enlarged, PMC may be a reasonable therapeutic option.8

The presence of thrombus in the LA is considered a general contraindication to this procedure. However, PMC can be performed if the thrombus is small fixed and located only in LA appendage under continuous transesophageal echocardiography guidance with acceptable risk in experienced hands.55

Pregnancy is a well-known trigger for the acute decompensation in RMS, and patients with severe MS are at increased risk for complications during pregnancy. Therefore, PMC is recommended before pregnancy for asymptomatic patients with severe MS who have valve morphology favourable for the procedure.8 ,39

For pregnant patients, PMC should be indicated only for those patients who remain symptomatic with NYHA class III to IV HF symptoms despite medical therapy as there is a risk of severe MR requiring urgent surgery.8 ,39 ,56 Several studies have shown the safety and efficacy of PMC on pregnant women especially when the procedure is performed after the 20th week, with improvement in the NYHA functional class as well as fetal results.57

Considering the progressive nature of rheumatic heart disease and the number of interventions in the past decades, the safety of reinterventions has been accessed in patients previously submitted to PMC. Data suggest that repeated procedures due to valve restenosis or failed PMC are safe in the long term, with similar event rates compared with elective intervention in a 25-year follow-up. In the largest published series, the 20-year reintervention rate was 38%. One-quarter of the patients underwent repeated PMC, with good immediate results in 94%, similar to the first intervention.58 Furthermore, compared with mitral valve surgery, PMC showed similar long-term event-free survival in a smaller cohort of reintervention.59 However, a prerequisite for considering repeat PMC for the treatment of restenosis is that restenosis is due to refusion of both commissures. PMC may not be feasible when restenosis is due to valve rigidity with persistent commissural opening.

Conclusions and perspectives

PMC is the treatment of choice for patients with symptomatic RMS. The cumulative experience with the technique and data from contemporaneous cohorts has demonstrated that PMC can be done safely in patients previously deemed to be unsuitable for this procedure. Newer scoring systems, which incorporate quantitative assessment of MV morphology, have optimised patient selection for PMC success. The determinants of PMV outcomes are multifactorial, mainly related to patient characteristics. Echocardiography plays a crucial role during procedural decision-making to have a more efficient procedure.

However, despite the advances over the past two decades there still remain unanswered questions regarding factors associated with long-term outcomes, especially for stable moderate MR, which are not adequately predictable by traditional echocardiographic variables. Expansion of the PMC indications should only be considered in centres with specialised expertise to ensure that this procedure is performed safely and effectively.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter Follow Bruno Nascimento at @ramosnas

  • Contributors All authors have contributed significantly to the paper.

  • Competing interests None declared.

  • Ethics approval Ethics approval was provided by the institutional clinical research and ethics committee.

  • Patient consent Obtained.

  • Provenance and peer review Commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.