Percutaneous mitral balloon commissurotomy in adults

INTRODUCTION — Mitral stenosis (MS) is a progressive disease that can lead to heart failure (most commonly presenting as dyspnea and sometimes causing fatigue and pulmonary edema) and additional severe complications (including systemic embolism, pulmonary hypertension, and death). Most cases of MS in adults are due to rheumatic heart disease, with symptoms appearing a variable number of years after one or more episodes of acute rheumatic fever, which are often undiagnosed. (See "Pathophysiology and natural history of mitral stenosis".)

Although medical therapy can relieve symptoms of MS, it does not affect the obstruction to flow. The interventions to treat MS are percutaneous mitral balloon commissurotomy (PMBC) [1-3] and mitral valve surgery (repair, commissurotomy, or valve replacement). (See "Surgical and investigational approaches to management of mitral stenosis" and "Rheumatic mitral stenosis: Overview of management", section on 'Indications for intervention'.)

Selection of candidates for PMBC, care of patients undergoing PMBC, complications, and outcomes will be reviewed here. The timing of mitral intervention (PMBC and mitral surgery) and medical and surgical management of MS are discussed separately. (See "Rheumatic mitral stenosis: Overview of management" and "Surgical and investigational approaches to management of mitral stenosis".)

USE — PMBC is predominantly used to treat rheumatic MS, although some types of congenital MS are also amenable to balloon dilation.

Rheumatic MS — The vast majority of patients undergoing PMBC have rheumatic MS. Commissural fusion is a key feature of rheumatic MS, and splitting of fused commissures is the mechanism of procedural success.

Indications for PMBC for rheumatic MS are discussed separately. (See "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'.)

Selective use for congenital MS — For most adults with congenital MS, PMBC is not a therapeutic option [4]. Parachute mitral valve is the most common form of congenital MS diagnosed de novo in adults [5], and MS caused by parachute mitral valve is generally not amenable to PMBC given the associated unbalanced chordal attachments [6].

This contrasts with the efficacy of PMBC in infants and children with congenital MS with typical congenital MS anatomy [7]. Among children diagnosed with congenital MS, parachute mitral valve is an uncommon cause.

Not used for MAC — Mitral annular calcification (MAC) commonly occurs in older adults and generally has no significant hemodynamic effects, but it can cause MS or mitral regurgitation. Annular calcification alone does not result in severe MS, but when calcification extends into the mitral inflow and leaflets are thickened and calcified, a gradient may be produced. MS caused by MAC is not amenable to PMBC. Management of MS caused by MAC is discussed separately. (See "Management and prognosis of mitral annular calcification", section on 'Mitral valve intervention'.)

ROLE OF HEART TEAM AND EXPERIENCED VALVE CENTERS — Evaluation and management of MS may be optimized by a heart team approach widely used for a variety of valve conditions [8,9]. This multidisciplinary approach incorporates clinical evaluation, scoring systems (see 'Scoring systems' below), and procedural risk assessment to determine if mitral valve intervention is indicated and to guide choice of intervention (PMBC or mitral surgery). (See 'Evaluation of candidates for PMBC' below and "Estimating the risk of valvular procedures" and "Preoperative evaluation for anesthesia for cardiac surgery".)

Cardiac surgical consultation is recommended even if PMBC is likely indicated, as in rare cases PMBC is complicated by the induction of severe mitral regurgitation resulting in acute hemodynamic instability requiring emergency mitral valve surgery. The patient should be informed of this infrequent but possible risk, and cardiac surgical back-up is recommended; it is optimal if the surgeon has already consulted on the case and the type of prosthetic valve that would be implanted has been clarified.

PMBC is most commonly performed in countries in which rheumatic fever is or has recently been prevalent; in such settings, procedure volume is often high and operator experience is often extensive. On the other hand, PMBC is infrequently performed in the United States and other countries in which acute rheumatic fever is rare; patients presenting with rheumatic MS in these countries with low prevalence of rheumatic fever are often immigrants from other countries where rheumatic fever persists [1]. Given the very low volume of patients with rheumatic MS treated by most clinical practices in the United States, professional societies have recommended that PMBC be performed at experienced valve centers (termed Comprehensive Valve Centers by the American College of Cardiology/American Heart Association [8,10] and Heart Valve Centres by the European Society of Cardiology [9]) where expertise and experience for low-volume procedures such as PMBC can be concentrated. In addition, the one US Food and Drug Administration (FDA)-approved system for PMBC, the Inoue balloon, is distributed by the manufacturer based on documented operator experience, and clinical representatives from the manufacturer are not generally present for the procedure (unlike procedures with more recently developed transcatheter valve therapies).

EVALUATION OF CANDIDATES FOR PMBC — Clinical assessment and echocardiography are used to evaluate potential candidates for PMBC [8,9,11-13]. (See "Rheumatic mitral stenosis: Clinical manifestations and diagnosis".)

Valve anatomy and function — Symptoms and echocardiography (primarily transthoracic echocardiography [TTE]) are used to stage MS (table 1), as well as any concomitant mitral regurgitation (MR) (table 2) and any other valve lesions. Transesophageal echocardiography (TEE) is particularly helpful in assessing the severity of MR, which may be underestimated on TTE due to acoustic shadowing.

Scoring systems

Rationale and limitations — Scoring systems based upon echocardiographic mitral features (valvular and subvalvular) are used to estimate the likelihood that a patient’s mitral valve will favorably respond to PMBC with hemodynamic and clinical improvement. Successful PMBC should reduce the severity of MS to a mild level and not produce moderate or greater MR. For patients who are candidates for both PMBC and cardiac surgery who have a high score (suggesting high risk of a suboptimal result from PMBC), cardiac surgery may be preferred. However, the choice of intervention is based upon the clinical context, including the patient’s surgical risk. (See "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'.)

PMBC is effective in splitting the fused mitral commissures caused by rheumatic heart disease, but the procedure does not address other causes of hemodynamic obstruction associated with rheumatic mitral valve disease, such as severe subvalvular disease and extensive thickening and rigidity of the valve leaflets. Echocardiography enables identification of features associated with suboptimal PMBC results, such as asymmetry of commissural fusion (ie, the two commissures fused to differing degrees), severe fibrosis and calcification of fused commissures, and severe subvalvular disease. In the setting of asymmetry of commissural fusion or calcification, the force produced by the inflating balloon may be transmitted to sites beyond the commissures, which may result in tearing of the anterior leaflet causing acute severe MR.

Various groups have developed criteria to predict the likelihood of PMBC success based on specific features in TTE and TEE studies. A role for intracardiac echocardiography in preprocedural assessment of subvalvular disease has been suggested but not established [14]. An ideal scoring system would incorporate multiple features that can be easily and reproducibly extracted from echocardiograms and combine these to yield a score that has been validated to identify patients likely to have successful PMBC outcomes, those patients likely to have adverse PMBC outcomes, and those with uncertain PMBC success.

Scoring systems are subject to the following limitations:

●Interobserver variability – Scoring systems require adequate image quality to enable qualitative assessments and quantitative measurements. These assessments and measurements are subject to interobserver variability, especially for observers not routinely performing these evaluations.

●Limited predictive value – Valve scores have limited predictive value. Some patients with unfavorable (high) scores may benefit from PMBC. Scores in the middle range may be less reliable than those at the extremes of the score spectrum. Even patients with optimum (low) scores are at risk for suboptimal outcomes. One reason for this is that some scores, such as the Wilkins score (see 'Wilkins score' below), predict the risk of insufficient valve opening by PMBC but do not address the risk of PMBC inducing MR. (See 'Postprocedural MR' below.)

Clinical decision-making requires consideration of clinical factors beyond the PMBC outcomes predicted by echocardiographic scoring systems. For example, PMBC may be an option for some patients with unfavorable (high) valve scores and high surgical risk. (See "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'.)

Choice of score — When evaluating candidates for PMBC, the chosen scoring system(s) should estimate the risk of an insufficient increase in valve area as well as the risk of induction of MR. All scoring systems provide estimates of outcomes and are not definitive. While the Wilkins score has been used for over three decades to predict PMBC outcomes, newer scoring systems (including the Nunes, Sutaria, and multifactorial scoring systems discussed below) have been proposed to improve predictive accuracy [15].

In our program we use a hybrid approach starting with the Wilkins score and then adding the presence of commissural fusion to provide a clinically useful assessment of the likelihood of gradient improvement and freedom from inducing severe MR. Some scoring systems require special measurements that we do not routinely perform.

Other options include using the Nunes score alone or combining the Wilkins score with assessment of commissural calcium and commissural area ratio (which is a component of the Nunes score) [15,16]. (See 'Wilkins score' below and 'Other scoring systems' below.)

Wilkins score — The Wilkins score was the first scoring system to predict the risk of insufficient improvement in valve opening with PMBC, and it continues to be the most commonly used score [12,15]. The Wilkins score is calculated from the sum of grades ranging from 0 (denoting the absence of the feature) to 4 (denoting severe abnormality) for each of four features of the mitral apparatus [12]:

●The degree of leaflet rigidity

●The severity of leaflet thickening

●The amount of leaflet calcification

●The extent of subvalvular thickening

Of note, the Wilkins score does not include a measure of the degree of commissural fusion or calcification or dense and thickened tissues associated with a fused commissure.

The maximum Wilkins score is 16; higher scores (>8) indicate more severe anatomic disease and a higher risk of suboptimal outcome from PMBC. A suboptimal outcome was defined as a postprocedure mitral valve area (MVA) <1.0 cm², left atrial pressure greater than 10 mmHg, and a less than 25 percent increase in MVA from baseline. The original study of the predictive value of the Wilkins score included 22 patients with total scores ranging from 4 to 16. There were 11 patients with scores between 7 and 11, of whom four had an optimal result and seven had a suboptimal result. These findings and later studies have demonstrated the clinical utility of the Wilkins score in candidates for PMBC as well as limitations of its predictive value for individual patients [15].

Other scoring systems

●Nunes score – The Nunes score was developed to enhance the accuracy of prediction of PMBC outcomes [17]. The Nunes score (unlike the Wilkins score) predicts the risk of inducing MR, as well as the risk of insufficient valve opening. This scoring system was developed using data from two cohorts of patients with severe MS: a 204-patient derivation cohort and a 121-patient validation cohort.

The Nunes score includes the following independent predictors of outcomes with assigned points based upon each feature’s regression coefficient:

•MVA ≤1 cm² – point value of 0 or 2.

•Maximum diastolic leaflet displacement ≤12 mm (assessed in the apical four-chamber view) – point value of 0 or 3.

•Commissural area ratio ≥1.25 (a measure of symmetry assessed in the parasternal short-axis view) – point value of 0 or 3.

and

•Subvalvular involvement – point value of 0 if absent or mild; point value of 3 if extensive thickening.

Three groups of patients with low (0 to 3), intermediate (5), and high (6 to 11) total scores had suboptimal short-term outcomes of 16.9, 56.3, and 73.8 percent. Similar stratification was observed in the validation cohort. This scoring system improved classification compared with the Wilkins score (net reclassification improvement 45.2 percent) The study also found that long-term outcomes were predicted by the patient’s age and three postprocedure variables: severity of MR, mean mitral gradient, and mean pulmonary artery pressure.

●Commissural calcium grade– A study in 149 patients of outcomes three years after PMBC found that presence of commissural calcification predicts adverse outcomes [18]. Survival at 36 months free of death, repeat PMBC, or mitral valve replacement (MVR) was significantly higher in patients without commissural calcium (86 vs. 40 percent) (figure 1) [18]. Commissural calcium was identified in five of six patients who developed severe MR following PMBC and underwent surgical MVR.

The Sutaria score was developed in a later study of 300 patients in which the extent of commissural calcification was quantified on a scale of 0 to 4 by giving each half of the posteromedial and anterolateral commissure a score of 0 or 1 [19]. Among patients with a Wilkins score ≤8, the likelihood of successful PMBC (achieving a MVA of >1.5 cm² without severe MR), was higher with low Sutaria score (grade of 0 or 1; 67 percent) than with a high Sutaria score (grade 2 or 3; 46 percent). An increase in MR of two or more grades occurred with similar frequencies in patients with low (score 0 or 1; 4.1 percent) or high (score 2 or 3; 2.8 percent) Sutaria scores.

●MR-risk score (Padial) – A scoring system to predict risk of PMBC-induced severe MR was developed that included scoring for leaflet thickening, commissural calcification, and subvalvular structure thickening [20,21]. In a series of 117 patients undergoing PMBC, those who developed severe MR had a significantly higher score compared with those who did not develop this complication (10.5 versus 8.2) [21]. When a total score of ≥10 was used as a cut-off, the sensitivity, specificity, and positive and negative predictive values for predicting severe MR were 82, 91, 90, and 97 percent, respectively.

This system was developed and reported from a single center's experience. It is the authors' experience that advanced valve deformity alone without commissural calcification does not predict a higher risk of inducing severe MR.

●Subvalvular apparatus (Bhalgat) score – A subvalvular apparatus (SVA) score may help predict adverse outcomes, although additional validation is needed. In a study from India of 356 patients with rheumatic MS undergoing PMBC in the course of one year, a scale of I to III was used to classify the degree of SVA disease [22]. The SVA score was graded I when neither of the two SVAs had severe disease, II when one of the two SVAs has severe disease, and III when both SVAs had severe disease. The echocardiographic findings were validated with surgical pathology specimens in 39 patients undergoing surgical valve replacement for either failure of PMBC to relieve obstruction or induction of severe MR. The SVA score was more sensitive and specific than the MR-risk score in predicting these adverse outcomes.

●Iung Method or Cormier Score – The results of echocardiography can be combined with other factors to improve patient selection. A logistic model was developed in a cohort of 1514 patients undergoing PMBC [23]. Predictors of good immediate outcomes (MVA ≥1.5 cm² with MR ≤2+) included younger age and less severe disease by echocardiography (including pliable leaflets with only mild chordal thickening, larger baseline MVA, and absence of fluoroscopic-detected calcification). Immediate results of PMBC were better with a large balloon only if there was no MR prior to the procedure.

CONTRAINDICATIONS — PMBC is contraindicated in patients with left atrial (LA) thrombus or moderate (2+) or greater mitral regurgitation. A suboptimal (high) valve score is a relative contraindication. (See 'Scoring systems' above.)

Left atrial thrombus

Identification — TEE is performed prior to PMBC to identify LA thrombus (most commonly seen in the LA appendage [LAA]), as it is more sensitive than TTE for this finding [8,9,24]. This difference was illustrated in a series of 474 patients undergoing mitral valve surgery in which the sensitivity and specificity of TTE for LA thrombi were 32 and 94 percent [25]. In contrast, the sensitivity and specificity of TEE for LAA thrombi were 98 and 98 percent, and for main LA cavity thrombi 81 and 99 percent. (See "Echocardiography in detection of cardiac and aortic sources of systemic embolism".)

The presence of LA thrombus is a contraindication to PMBC because of the risk of dislodging the thrombus during the procedure. Such thrombi are not uncommon in patients with MS. The prevalence is related in part to whether or not the patients have been treated with warfarin. Thus, in two series of patients referred for PMBC, the prevalence of LA thrombus was 13 percent in a study from Scotland in which almost all patients had been treated with warfarin [26] compared with 32 percent in a report from Thailand, which excluded patients who had been treated with warfarin more than 72 hours before study entry [27].

Management — If LA thrombus is detected (generally by TEE) in a potential candidate for PMBC, the procedure is postponed and anticoagulant therapy is administered. For most patients in this setting, we treat with vitamin K antagonist (VKA; eg, warfarin) with target international normalized ratio (INR) 3.0, range 2.5 to 3.5. Thrombus resolution is assessed by repeat TEE after six months of anticoagulation. Some patients with LA thrombus found on TEE immediately prior to planned PMBC are too symptomatic to wait six months. In such patients, repeat TEE may be performed after a shorter duration of anticoagulation (eg, one month).

While improved imaging during PMBC, as with real-time three-dimensional (3D) TEE, may enable more careful catheter manipulation in the LA, it is not possible to ensure that the catheter tip will not end up unintentionally in the LAA during some part of the procedure. Given the risk of embolic stroke from an LA thrombus, PMBC is contraindicated when LA thrombus is present.

For patients with MS with an indication for mitral valve intervention and LA thrombus that does not resolve with anticoagulant therapy, PMBC should not be performed; instead, mitral valve surgery (repair, commissurotomy, or replacement with LAA ligation) is performed if surgical risk is not prohibitive. Surgical mitral valve repair or replacement with LAA ligation may also be considered in patients unable to tolerate an adequate duration of anticoagulation. (See "Rheumatic mitral stenosis: Overview of management", section on 'Our approach' and "Atrial fibrillation: Left atrial appendage occlusion", section on 'Patients undergoing surgery'.)

Outcomes with VKA — The frequency of thrombus resolution was assessed in a report of 219 patients from Thailand [27]. Oral anticoagulation was initiated to maintain an INR of 2.0 to 3.0. At six months, thrombus was no longer detectable in 53 patients (24 percent). Variables predicting thrombus resolution included better New York Heart Association functional class (I or II), thrombus size ≤1.6 cm² and confined to the LAA, less marked LA spontaneous echocardiographic contrast, and maintenance of a mean INR ≥2.5. Patients with all these predictors had a 94 percent incidence of complete thrombus resolution.

In a study from India of 66 patients with MS and LA thrombus, oral anticoagulation was given for up to six months [28]. Complete resolution of thrombus was seen on TEE in 22 patients (33 percent) and organization in 38 (58 percent); only six patients (9 percent) had no change in thrombus appearance.

Observational data on VKA therapy in patients with LA thrombus without MS suggest that thrombus resolution is faster and more complete than in patients with MS. Two observational trials initiated VKA therapy for four weeks in patients with nonvalvular atrial fibrillation (AF) found to have atrial thrombus on TEE. In one study, 11 patients were treated with four weeks of warfarin. Repeat TEE showed thrombus resolution in 9 of 11 patients (81.8 percent) [29]. In a later study, 14 patients with nonrheumatic AF not on anticoagulants presented with atrial thrombus (LAA thrombus in 14, LA body thrombus in two, right atrial appendage thrombus in one, and right atrial body thrombus in one). After warfarin anticoagulation with a median duration of four weeks and with an INR goal 2.0 to 2.8, repeat TEE showed 16 of 18 atrial thrombi (89 percent) had resolved [30].

Outcomes with DOAC — Scant data are available on the use of direct oral anticoagulants (DOAC) in the setting of rheumatic MS with TEE-detected LA thrombus. A case report described resolution of a large LA thrombus in a patient with MS after five weeks of therapy with rivaroxaban. There are case reports of the use of DOACs for LA thrombus resolution in patients with nonvalvular AF [31]. A survey of European centers performing cardioversion for AF revealed a diversity of approaches when thrombus is detected [32]. If the patient had not been on anticoagulation, 30 percent would use a VKA while 64 percent would use a DOAC. In patients with persistent thrombus despite either VKA or DOAC, most centers would change to a different anticoagulant. Repeat TEE imaging was commonly recommended to assess all approaches to resolve LA thrombus.

The use of rivaroxaban for newly diagnosed LA thrombus in the setting of nonvalvular AF was the subject of the X-TRA single-arm prospective study. Of the 53 study subjects, 75 percent had not received anticoagulation therapy and the remainder had been on subtherapeutic VKA when TEE revealed a LAA thrombus. TEE was repeated after six weeks of rivaroxaban with documentation of thrombus resolution in 41.5 percent, reduced thrombus size in 19 percent, unchanged thrombus size in 17 percent, and increased thrombus size in 22.5 percent [33].

THE PROCEDURE — PMBC should be performed in an experienced valve centers (termed Comprehensive Valve Centers [8,10] or Heart Valve Centres [9]) with cardiac surgical backup, as discussed above. (See 'Role of heart team and experienced valve centers' above.)

Technique — PMBC is typically performed via right femoral venous access with transseptal puncture to gain access to the left atrium and mitral valve. The FDA-approved Inoue system comes in three different balloon sizes, which are matched to the patient's height and weight. A single deflated Inoue balloon catheter (or a double-balloon system) is advanced from the venous circulation to the right atrium, across the interatrial septum via transseptal puncture to the left atrium, and finally across the stenotic mitral valve. Inflation and rapid deflation of the balloon opens the stenotic valve via separation of the fused commissures (movie 1). This mechanism is similar to that of surgical commissurotomy [34]. Fracture of calcific deposits in the leaflet tissue may also contribute to improved leaflet mobility and leaflet separation. More than one inflation is often needed, and the Inoue balloon system allows serial inflations with increasing fluid volumes to inflate the balloon.

When PMBC is performed in patients with high or prohibitive cardiac surgical risk, sizing of the balloon and the degree of each inflation should be conservative, and a moderate reduction in MS severity should be accepted given the lack of surgical option if severe mitral regurgitation (MR) is induced. (See "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'.)

Monitoring — The standard approach to monitoring PMBC includes direct measurement of the transmitral pressure gradient and echocardiography with measurement of transmitral gradient using Doppler, mitral valve area (MVA) using planimetry, assessment of commissural splitting, and assessment of MR severity after each balloon dilation. The procedure is stopped when an adequate valve area has been achieved or if there is a significant increase in MR severity.

Use of TTE, TEE, or intracardiac echocardiography (ICE) during PMBC is influenced by local resources and experience. TTE can be used to assess the results after each inflation but is not useful in performing the initial transseptal catheterization.

TEE can be performed during PMBC to guide the transseptal puncture, the PMBC procedure, and to monitor for complications [13,35]. TEE allows for direct assessment of the degree to which fused commissures have been split. Real-time 3D TEE is used in most centers to guide PMBC. This enables improved visualization of catheter manipulations, a more comprehensive assessment of the mitral valve before and after PMBC, and a more visually intuitive guidance. 3D TEE may identify commissural splitting and leaflet tears that are not visible on a two-dimensional study. Improved outcomes from a 3D approach have not yet been reported. However, TEE has the disadvantage that general anesthesia is generally required. (See "Three-dimensional echocardiography", section on 'Mitral valve'.)

ICE (figure 2) can delineate the extent of valvular deformity (image 1 and image 2) and may also be useful in visualizing several key steps of the PMBC procedure, including [14]:

●Transseptal puncture of the interatrial septum (image 3)

●Exclusion of thrombus from the left atrial appendage

●Optimizing balloon placement across the valve orifice (image 4)

●Initial assessment of the results of dilation

Disadvantages of ICE are increased cost for some centers and, with available catheters and transducers, an inability to accurately measure MVA, difficulty in assessing transmitral Doppler gradients, and inadequate assessment of MR.

OUTCOMES

Short-term hemodynamic effects — PMBC generally provides effective immediate relief of mitral valve obstruction in appropriately selected patients. Most investigators report that the valve area increases from 1.0 cm² or less to 2.0 cm² or larger, a change that is associated with a rapid reduction in left atrial (LA) pressure (figure 3). This result is often smaller than the normal valve area of 3 to 5 cm², but it is sufficient to produce substantial hemodynamic improvements including a decrease in LA pressure, the transmitral pressure gradient, and pulmonary artery pressure, plus an increase in cardiac output [1]. An additional benefit is a reduction in LA stiffness, resulting in an increase in LA pump function in patients in sinus rhythm and an increase in LA reservoir function in those with AF [36].

Complications — The risk of complications for elective PMBC depends upon patient selection and operator experience and generally compares favorably with complication rates with surgical commissurotomy. The following complication rates have been reported for elective PMBC: mortality ≤0.5 percent, embolic events caused by air or thromboembolism (including stroke) 0 to 4 percent, cardiac tamponade 0 to 2 percent, and induction of mitral regurgitation (MR; new or worsening) 1.4 to 9.1 percent [37-40]. The effect of postprocedural MR on long-term outcomes is discussed below. (See 'Postprocedural MR' below.)

In contrast, the mortality rate is high when emergency PMBC is performed in patients with cardiogenic shock, cardiac arrest, or pulmonary edema refractory to medical therapy [41].

The transseptal puncture performed during the PMBC procedure commonly causes a trivial interatrial shunt [15,42,43], and hemodynamically significant shunts (Qp:Qs ≥1.5:1) are rare [44]. A residual iatrogenic atrial septal defect is identified in up to 20 percent of patients after one year [45].

Long-term outcomes — Long-term outcomes are impacted by progressive reduction in mitral valve area (MVA). Twelve months after valvotomy, several hemodynamic parameters may show continued improvement, including further reductions in pulmonary artery systolic pressure and increases in cardiac output. Pulmonary vascular resistance declines and normalizes in many patients [46]. However, in a long-term follow-up of 561 patients, there was a progressive reduction in the MVA which, by seven years, was ≥0.3 cm² in 27 percent of patients [38]. In comparison, mild to moderate MR after the procedure did not progress in 81 percent and, when progression occurred, it was generally by one grade. (See 'Postprocedural MR' below.)

A review from the United States National Heart, Lung, and Blood Institute (NHLBI) Balloon Valvuloplasty Registry evaluated 736 patients over the age of 18 who were followed for four years [47]:

●The actuarial survival rate at one, two, three, and four years was 93, 90, 87, and 84 percent, respectively.

●The event-free survival (freedom from death, mitral valve surgery, or repeat PMBC) at one, two, three, and four years was 80, 71, 66, and 60 percent, respectively.

The need for repeat intervention, with either surgery or PMBC, is more common as time passes. Among a cohort of 912 patients with good immediate results following PMBC (MVA ≥1.5 cm² with less than moderate MR) who were followed for a median of 12 years, 38 percent required reintervention (266 surgery, 85 repeat PMBC) [48]. Overall survival for the cohort was 75 percent at 20 years, with 46 percent cardiovascular survival free of mitral surgery at 20 years.

Patients with MS and pulmonary hypertension may develop functional tricuspid regurgitation. Successful PMBC with decrease in pulmonary artery pressures is associated with improvement in tricuspid regurgitation in some but not all patients, and some patients may require tricuspid valve surgery. (See "Management and prognosis of tricuspid regurgitation" and "Rheumatic mitral stenosis: Overview of management", section on 'Concomitant indication for cardiac surgery'.)

Predictors of long-term outcome — The long-term results depend upon baseline clinical, hemodynamic, and echocardiographic features in addition to the acute result from PMBC [38,41,47,49-54].

Among patients treated electively, independent predictors of longer event-free survival include less deformed valves, normal ventricular function, and better baseline New York Heart Association (NYHA) functional class (table 3) [49]. In a review of 146 patients, those who fulfilled two or three of these criteria had five-year event-free survival rates of 60 to 84 percent, respectively, compared with 13 to 41 percent who fulfilled none or only one of these criteria [49].

These predictors have been confirmed by larger studies [38,47,51-54]. In the above cited NHLBI review, for example, independent predictors of mortality were functional NYHA class IV (table 3), higher Wilkins echocardiography deformity score (see 'Wilkins score' above), and higher postprocedural pulmonary artery systolic and left ventricular end-diastolic pressures [47].

Similar findings were noted in a report of 879 patients who were followed for 12 years [53]. Regression analysis identified the following as independent predictors of adverse events (death, mitral valve surgery, and repeat PMBC):

●Wilkins score >8, which was associated with lower success rates and smaller increases in MVA (see 'Wilkins score' above)

●Increasing age

●Prior surgical commissurotomy

●NYHA functional class IV

●Higher postprocedural pulmonary artery pressure

●Preprocedural MR ≥2+

●Postprocedural MR ≥3+

Patients with a Wilkins score ≤8 had, at a follow-up of 12 years, significantly higher rates of patient survival (82 versus 57 percent in patients with a higher score) and total event-free survival (38 versus 22 percent).

A later study from South Korea included 742 patients who had routine follow-up for more than 10 years after PMBC [55]. An optimal result, as defined by postprocedure MVA >1.5 cm² and MR grade ≤2+, was achieved in 85 percent of patients with an average age of 41 years. During the subsequent follow-up (mean of 214 months), 7.3 percent underwent repeat PMBC, 0.5 percent underwent two additional PMBC procedures, and 33.4 percent underwent surgery; 4.7 percent died from cardiovascular causes, and 4.4 percent had a stroke. After adjusting for confounding variables, Wilkins score (see 'Wilkins score' above) and postprocedure MVA were the only predictors of long-term outcomes.

Differences in the prevalence of these predictive clinical factors likely explains observed differences in clinical outcomes for various patient populations in different countries. Studies from countries with prevalent acute rheumatic fever typically involve younger patients with lower echocardiographic scores, who therefore may have event-free survival rates at 5 to 10 years of well over 75 percent (comparable with that seen with surgical commissurotomy). In contrast, North American patient series tend to involve older patients with more severe valvular deformity and calcification and, consequently, a much lower event-free survival rate [1].

Atrial fibrillation — Many patients with MS have AF, which can adversely affect the short- and long-term outcome after surgical mitral commissurotomy [56]. The impact of AF on the results after PMBC was addressed in a study that compared the outcome of 355 patients with AF with 379 in normal sinus rhythm [57]. The following results were noted:

●Patients with AF were older and more frequently presented with NYHA class IV heart failure, Wilkins score >8, calcified valves, and a history of previous commissurotomy.

●Procedural success was lower when AF was present (61 versus 76 percent with sinus rhythm), and in-hospital mortality was higher (3 versus 0.5 percent).

●The long-term outcome was poorer in those with AF compared with those with sinus rhythm, largely due to a smaller postprocedure MVA (1.7 versus 2.0 cm²); after a 60-month follow-up, overall survival (68 versus 89 percent) and event-free survival (without death, repeat valvotomy, or mitral valve surgery; 32 versus 61 percent) were lower in the patients with AF.

●In those with AF, independent predictors of events were similar to those described above: severe postprocedure MR (≥3+), Wilkins score >8, and preprocedural NYHA class IV heart failure.

Another report evaluated the efficacy of electrical cardioversion one month after PMBC in 71 patients with MS who had AF [58]. Cardioversion was successful in 50 (71 percent); these patients were treated with amiodarone (200 mg/day) until the end of the study at one year. AF recurred in 24 (48 percent), and increased LAdiameter was the only predictor of recurrence. The outcomes were best in the 43 percent of patients who had an LA diameter <60 mm and no associated aortic valve disease; approximately two-thirds of these patients were in sinus rhythm at one year.

Postprocedural MR — Induction or worsening of MR is a complication of PMBC but is rarely severe enough to require treatment with emergency mitral valve replacement (MVR). More commonly, MR is mild to moderate, which may affect the longer-term need for MVR [38,39]. The frequency, causes, and natural history of significant postprocedural MR was evaluated in a follow-up study of 380 patients who underwent PMBC [39]. The following findings were noted:

●Significant MR (more than mild) developed in 12.4 percent of patients.

●The most frequent mechanism (57 percent) was MR that originated at the site of successful commissurotomy. Noncommissural MR resulted from leaflet laceration or subchordal damage.

●The survival rate at eight years was not statistically different between those with and without MR (96 and 98 percent, respectively).

●The eight-year event-free survival (absence of cardiac death, MVR, second PMBC, heart failure requiring admission, or embolism) rate was significantly lower in patients with significant MR than in those without MR (48 versus 83 percent, respectively). Patients with commissural MR had a higher rate of event-free survival than those with noncommissural MR (63 versus 29 percent). The frequency of surgical MVR in patients with commissural and noncommissural MR was 15 and 70 percent, respectively.

Older adults — The long-term event-free survival after PMBC may be lower in older adults, largely due to more comorbidities (including coronary heart disease) and advanced valve deformity and calcification [1,47,59]. Nevertheless, outcomes may still be satisfactory, particularly in patients with favorable valve morphology [47,59-61]. (See "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'.)

As discussed separately (see "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'), PMBC is also an effective, low-risk palliative treatment in older adults who have unfavorable valve morphology (marked valvular and subvalvular degenerative changes that would normally require valve replacement), but are high risk candidates for surgery [60]. The technique is adjusted in this clinical setting as described above. (See 'Technique' above.)

For PMBC post-surgical commissurotomy — PMBC has been performed in patients who have developed mitral valve restenosis after previous surgical commissurotomy. The immediate hemodynamic results and the long-term hemodynamics and outcome at 12 to 40 months appear to be similar to those seen with PMBC in unoperated MS [62-64].

As an example, one report described 232 patients who underwent PMBC a mean of 16 years after surgical commissurotomy: 82 percent had good immediate results (MVA ≥1.5 cm² without more than moderate MR); one patient died (0.4 percent); and 4 percent had greater than moderate (2+) MR [64]. After an eight-year follow-up, 58 percent of those with good immediate results were in NYHA class I or II (table 3).

Of note, patients selected for PMBC with restenosis after surgical commissurotomy are a subset often without advanced deformity of the valve apparatus and without significant MR. Many patients with restenosis after surgical commissurotomy are further along in the chronic scarring process and some have other valvular disease requiring surgical intervention. Such patients generally require surgical MVR rather than PMBC.

POST-PMBC CARE

Follow-up — While successful PMBC may postpone mitral valve surgery for as long as a decade or more, restenosis requiring intervention (eg, repeat PMBC or mitral valve replacement) is likely to be necessary in the long term. Monitoring is recommended to assess for symptoms and signs of progressive mitral valve obstruction and to evaluate any concomitant valve disease. In older adults, restenosis is usually due to superimposed fibrosis and calcification of the valve leaflets rather than recurrent commissural fusion. In contrast, restenosis in younger adults may be due to recurrent rheumatic fever with recurrent commissural fusion; appropriate secondary prophylaxis against rheumatic fever is important in these patients. (See "Rheumatic mitral stenosis: Overview of management", section on 'Monitoring' and "Rheumatic mitral stenosis: Overview of management", section on 'Secondary prevention of rheumatic fever' and "Acute rheumatic fever: Treatment and prevention" and "Acute rheumatic fever: Treatment and prevention", section on 'Secondary prevention (antibiotic prophylaxis)'.)

Repeat PMBC — In patients who develop indications for intervention after PMBC, options include mitral valve surgery and repeat PMBC. As for primary PMBC, repeat PMBC is favored in patients with optimal (low) echocardiographic scores and/or at high risk for complications with mitral valve surgery [65-67].

In a series of 36 patients with symptomatic MS who underwent repeat PMBC, the following results were noted [66]:

●The immediate procedural success rate was 75 percent.

●At one and three years, overall survival was 74 and 71 percent, respectively, and event-free survival was 61 and 47 percent, respectively.

●Patients who had comorbid cardiac or noncardiac diseases had a much lower two-year event-free survival compared with those without another disease (29 versus 86 percent) (figure 4).

●Independent risk factors for event-free survival after repeat PMBC were the same as those for the first procedure and included preprocedural echocardiographic score and postprocedural MVA, mitral regurgitation (MR) severity, and pulmonary artery pressure.

Exercise — Among patients who have undergone successful PMBC, participation in sports should, as in patients with untreated lesions, be based upon the residual severity of MS and MR and the possible presence of left ventricular dysfunction [68]. The capacity to exercise should be evaluated with an exercise test to at least the level of anticipated activity. (See "Chronic primary mitral regurgitation: General management", section on 'Exercise' and "Rheumatic mitral stenosis: Overview of management", section on 'Physical activity and exercise'.)

Pregnancy — The management of MS (including PMBC) prior to, during, and following pregnancy is discussed separately. (See "Pregnancy in women with mitral stenosis".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cardiac valve disease".)

SUMMARY AND RECOMMENDATIONS

●Use – Interventions to treat mitral stenosis (MS) include percutaneous mitral balloon commissurotomy (PMBC) and mitral valve surgery (repair, commissurotomy, or valve replacement). PMBC is predominantly used to treat rheumatic MS, although some types of congenital MS are also amenable to balloon dilation. MS caused by mitral annular calcification (MAC) is not amenable to PMBC. (See 'Use' above.)

Specific recommendations for use of PMBC in patients with rheumatic MS are provided separately. (See "Rheumatic mitral stenosis: Overview of management", section on 'Indications for intervention'.)

●Role of experienced valve centers – Experienced valve centers (known as Comprehensive Valve Centers or Heart Valve Centres) serve as referral centers for concentration of expertise and experience for management of MS, particularly PMBC. (See "Rheumatic mitral stenosis: Overview of management", section on 'Role of heart team and experienced valve centers'.)

●Scoring systems – Scoring systems based upon echocardiographic mitral features (valvular and subvalvular) are used to estimate the likelihood that a patient’s mitral valve will favorably respond to PMBC with hemodynamic and clinical improvement. However, all scoring systems provide estimates of outcome and are not definitive, and other clinical factors often come into play in considering PMBC. (See 'Scoring systems' above.)

The chosen scoring system(s) should estimate the risk of insufficient improvement in mitral valve area (MVA) as well as the risk of induction of mitral regurgitation (MR). We use a hybrid approach starting with the Wilkins score and then adding the presence of commissural fusion. (See 'Choice of score' above.)

●Contraindications – PMBC is contraindicated in patients with left atrial (LA) thrombus or moderate (2+) or greater MR. A suboptimal echocardiographic score is a relative contraindication. (See 'Contraindications' above and 'Scoring systems' above.)

●Management of LA thrombus – If LA thrombus is detected by transesophageal echocardiography (TEE) in a potential candidate for PMBC, the procedure should be postponed while anticoagulant therapy is administered, as discussed in greater detail separately. (See 'Left atrial thrombus' above and "Rheumatic mitral stenosis: Overview of management", section on 'Prevention of thromboembolism'.)

A follow-up TEE is performed after six months of anticoagulation (sooner if the patient is too symptomatic to wait six months). If LA thrombus does not resolve with anticoagulation, the patient is referred for mitral valve surgery with LA appendage (LAA) ligation. (See "Atrial fibrillation: Left atrial appendage occlusion" and "Rheumatic mitral stenosis: Overview of management", section on 'Our approach'.)

●The procedure – PMBC is typically performed via right femoral venous access and transseptal puncture to access the stenotic mitral valve with hemodynamic and echocardiographic monitoring. Inflation and rapid deflation of the balloon opens the stenotic valve via separation of the fused commissures (movie 1). (See 'The procedure' above.)

●Complications – The risk of complications for PMBC depends upon patient selection and operator experience and generally compares favorably with complication rates with surgical commissurotomy. The following complication rates have been reported for elective PBMC: mortality ≤0.5 percent, embolic events caused by air or thromboembolism (including stroke) 0 to 4 percent, cardiac tamponade 0 to 2 percent, and induction of MR (new or worsening) 1.4 to 9.1 percent. The transseptal puncture commonly causes a trivial interatrial shunt, and hemodynamically significant shunts are rare. (See 'Complications' above and 'Postprocedural MR' above.)

●Outcomes – PMBC generally provides effective immediate relief of mitral valve obstruction in appropriately selected patients. Long-term outcomes are impacted by progressive reduction in MVA, which result in need for reintervention (in approximately 40 percent at 5- to 10-year follow-up, but rates vary widely). (See 'Outcomes' above.)

●Post-PMBC care – While successful PMBC may postpone mitral valve surgery for a decade or more, restenosis requiring intervention (eg, repeat PMBC or mitral valve replacement) is likely to be necessary in the long term. Monitoring is recommended to assess for development of indications for reintervention (repeat PMBC or mitral valve surgery). (See 'Post-PMBC care' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges William H Gaasch, MD (deceased), who contributed to earlier versions of this topic review.