Currently Used Scoring System

Wilkins et al devised the echocardiographic scoring system that is currently the most widely used. Leaflet thickness, leaflet mobility, leaflet calcification, and the extent of subvalvular disease are each evaluated and scored on a scale of 1 to 4. The sum of the 4 scores is a predictor of the likelihood of successfully opening the stenotic valve. However, recent studies have questioned the precision of this score as a predictor of outcomes and have suggested the need for a more refined echocardiographic assessment. For example, several authors have emphasized careful evaluation of the degree of commissural fusion and calcification as critical for predicting the outcome of PBMV. Importantly, these observations are not included in the Wilkins score. However, the importance of assessment of the commissures has so far been validated only in small numbers of patients. More complete testing and comparison with more widely accepted scoring system are required. Table 1 lists some of the limitations of the scoring systems based on the mitral apparatus morphology, in particular the Wilkins score system.

Potential Advantages of the New Scoring System

In this issue, Anwar et al propose a new score for selecting patients for PBMV using real-time three-dimensional (3D) echocardiography (RT3DE). Their novel score addresses several of the shortcomings of the Wilkins score ( Table 1 ). First, by separately evaluating each segment of both the anterior and the posterior mitral leaflets (a total of 6 segments), they account for any uneven distribution of anatomic abnormalities in patients with rheumatic MS. Second, their score includes assessment of the commissural areas. Third, they sought to identify the relative importance of each parameter to a prediction of the likelihood of a successful PMBV. With that in mind, they have given more or less weight to the individual scores.

Because the investigators aimed to compare their system with that of Wilkins et al, it is important to note that they modified the Wilkins score by using a grading system of 0 to 4 instead of the conventionally used grading system of 1 to 4. This change in grading automatically lowered their assessment of mitral valve anatomy by the Wilkins score and somewhat “muddies” (taints) their conclusion that the RT3DE score tends to be higher and less favorable.

Potential Advantages of the New Scoring System

In this issue, Anwar et al propose a new score for selecting patients for PBMV using real-time three-dimensional (3D) echocardiography (RT3DE). Their novel score addresses several of the shortcomings of the Wilkins score ( Table 1 ). First, by separately evaluating each segment of both the anterior and the posterior mitral leaflets (a total of 6 segments), they account for any uneven distribution of anatomic abnormalities in patients with rheumatic MS. Second, their score includes assessment of the commissural areas. Third, they sought to identify the relative importance of each parameter to a prediction of the likelihood of a successful PMBV. With that in mind, they have given more or less weight to the individual scores.

Because the investigators aimed to compare their system with that of Wilkins et al, it is important to note that they modified the Wilkins score by using a grading system of 0 to 4 instead of the conventionally used grading system of 1 to 4. This change in grading automatically lowered their assessment of mitral valve anatomy by the Wilkins score and somewhat “muddies” (taints) their conclusion that the RT3DE score tends to be higher and less favorable.

Potential Limitations of the New Scoring System

It should be recognized that use of 3D echocardiography for the purpose of assessing the mitral apparatus in patients with rheumatic MS has limitations. Several examples can be cited. First, just as is true with two-dimensional (2D) echocardiography, 3D echocardiography has not been validated for differentiating nodular fibrosis from calcification. Both nodular fibrosis and calcium deposition in leaflets can cause increased thickness and reflectance, but splitting the former is likely to be easier than splitting the latter during PBMV. Second, RT3DE currently has poorer resolution than 2D echocardiography. Thus, structures appear thicker than they actually are. Anwar et al assert that the determination of leaflet thickness is easier by 3D compared with 2D echocardiography and point out that 76% of their patients had high thickness scores, compared with 60% on 2D echocardiography. However, they fail to point out the limitation imposed by the difference in resolution. Therefore, the utility of this particular finding needs further substantiation. Third, the authors also assert that RT3DE is superior for evaluating the subvalve apparatus. However, they compared RD3DE using multiple views with a single view (the apical 4-chamber view) by 2D echocardiography. We, and many others, evaluate the subvalve apparatus using multiple views. We include off-axis parasternal long-axis views, tilting the transducer both medially and laterally to evaluate both papillary muscle–chordae systems, which are not imaged in a standard parasternal long-axis view. We believe this series of views to be superior to an apical 4-chamber view.

Interestingly, the investigators graded the degree of mitral regurgitation solely by the vena contracta method. Although often used for the purpose, the vena contracta has known limitations and generally should not be used as the sole criterion for quantitating mitral regurgitation. Moreover, the use of vena contracta to evaluate mitral regurgitation has not been validated in patients with MS, in whom the anatomy of the regurgitant orifice is somewhat unique relative to the majority of other etiologies of mitral regurgitation.

We applaud Anwar et al for using newer technology in an attempt to overcome some of the shortcomings of the commonly used Wilkins score system. We also endorse the need for an improved scoring system. We agree with the authors that transesophageal RT3DE is likely to be proved complementary to currently used scoring methods. Moreover, one can anticipate that 3D technology, especially resolution, will continue to improve. Unfortunately, the current lack of widespread availability will delay the rapid adoption of RT3DE.

Should an increase in mitral valve area be the sole endpoint for scoring system? We believe that there is an even more fundamental issue that is not addressed in any current scoring system. To date, all systems define “success” as the attainment of a specific goal for mitral valve area. However, in our view, PBMV should not be restricted to patients who have a high probability of improvement in valve area on the basis of a low score using any scoring system or combination of systems. To do so could lead to the exclusion of many patients who may also achieve good results with acceptable safety. In fact, good immediate and long-term results have been reported after PBMV in patients with “unsuitable” valve anatomy, on the basis of current scoring systems. Post et al identified 72 patients with Wilkins scores ≥ 10 who underwent PBMV. Thirty-two of these 72 patients (44%) had event-free survival and were in functional class I or II at 3 years. Abascal et al observed that 32% of patients with scores > 8 had good immediate results.

At our institution, with a large experience with PBMV, we choose it as an initial treatment modality in the vast majority of patients with symptomatic MS, almost without regard to score. We reserve surgery for instances in which valve opening has been unsuccessful or in patients who develop severe mitral regurgitation associated with clinical deterioration. The development of significant mitral regurgitation is generally regarded as a failure of PBMV, requiring surgery. However, in our experience, and that of others, even severe mitral regurgitation resulting from PBMV is well tolerated by the majority of patients, and urgent surgery is seldom required for this complication. Clinical tolerance to immediately produced severe MR is variable. Although a few require emergency surgery, the majority tolerate the sudden MR, allowing planned surgery days to weeks later, while some patients experience decreases in MR severity over time and do not require surgery. In one study, 1 in 5 patients went on to have surgery after PBMV. Only a small proportion (18%) of these required surgery within 2 weeks. Failed or complicated PBMV was not catastrophic, and the vast majority (82%) of patients had elective surgery >2 weeks later. The reason that this sudden increase in degree of mitral regurgitation is well tolerated may relate to the fact that these patients typically have at least moderately dilated and compliant left atria and have already adapted to chronically elevated left atrial pressures.

The strategy of a trial of PBMV, even in patients with high scores, is also often rewarded by good and durable results. When the risk of surgery is high, such as in elderly patients, this approach is especially rewarding.