The introduction of percutaneous balloon mitral valvuloplasty (BMV) by Inoue et al¹ in 1982 has opened a new dimension in the treatment of patients with mitral stenosis. The body of data accrued to date has clearly established this invasive, nonsurgical procedure as the treatment of choice in symptomatic patients with moderate to severe mitral stenosis (mitral valve area <1.5 cm²) and favorable valve morphology (noncalcified, pliable valve with minimal subvalvular disease and no or mild mitral regurgitation).^2-5 The presence of either severe (grade ≥3) angiographic mitral regurgitation or left atrial thrombus is considered to be a contraindication for BMV. However, several controversial issues in the use of this procedure exist, because the selection of patients for BMV in clinical practice continues to be a complex decision involving consideration of multiple variables, including clinical profile, operator skill, valve morphology, and severity of associated mitral regurgitation.

Among the multiple variables, valve morphology and severity of associated mitral regurgitation have, to a large extent, remained the principal determinants in patient selection. In patients with favorable valve morphology (pliable, noncalcified valve with minimal subvalvular disease) and no or mild mitral regurgitation, BMV predictably yields excellent results and a low risk of resultant severe mitral regurgitation. With successful balloon valve enlargement, there is generally a 2-fold increase in the mitral valve area and an associated dramatic fall in transmitral valve gradient, left atrial pressure, and pulmonary artery pressure.^2-5 These hemodynamic benefits are mirrored in clinical improvements in the patients’ symptoms and exercise tolerance.

This is not entirely surprising, considering the fact that BMV enlarges the stenosed mitral valve in the same manner as that afforded by surgical commissurotomy—namely that of commissural split. Several randomized trials^6-11 comparing BMV and closed and/or open surgical commissurotomy in patients with favorable valve morphology have demonstrated that BMV is as efficacious, if not more so, than surgical mitral commissurotomy in acutely relieving the obstructed valve and achieving favorable clinical outcome. The long-term results of BMV are excellent, especially when the acute results are optimal and valve morphology is good. Long-term data in BMV have also indicated that after optimal mitral valve dilation, the restenosis rate is low and the acute symptomatic benefits are sustained.^12-16 When restenosis is defined as mitral valve area less than 1.5 cm², the restenosis rate in patients with favorable valve morphology at 7-year follow-up in the randomized trial by Farhat et al¹¹ was 6.6% in patients who underwent BMV; this rate was similar in those who underwent open surgical commissurotomy and was far superior to the restenosis rate of 37% observed after closed surgical commissurotomy. Hernandez and associates¹⁶ found that survival free of major events (cardiac death, mitral surgery, repeat BMV, or functional impairment) was 69% at 7 years, ranging from 88% to 40% in different subgroups of patients. Mitral area loss, although mild (0.13 ± 0.21 cm²), increased with time and was greater than or equal to 0.3 cm² in 12%, 22%, and 27% of patients at 3, 5, and 7 years, respectively.

In contrast to the usually excellent results obtained with the procedure in patients with favorable valve anatomy, the results of BMV in those with adverse valve morphology (heavily calcified leaflets and commissures and extensive subvalvular disease) are less predictable.^17-20 BMV in the latter compared with the former patient subset tends to produce inferior acute and long-term results. Therefore, patients with unfavorable mitral morphology are, in general, better served with mitral valve replacement (but not surgical commissurotomy). Having said that, there remains a significant minority of patients with adverse valve morphology in whom BMV may yield acceptable hemodynamic outcome and may continue to gain long-term symptomatic improvement. It is thus reasonable to perform BMV in these patients if they refuse surgery as a bridge to surgery at a later stage, particularly if urgent noncardiac surgery is required or if the risk of cardiac surgery is deemed to be prohibitively high because of other major medical comorbidities (Table 41-1). In fact, BMV may occasionally be the only option for some of these patients.

Many centers performing BMV exclude patients with moderate (angiographic grade 2+) mitral regurgitation from the procedure for fear of increasing the severity of the mitral regurgitation and the need for emergency mitral valve surgery. In our centers, these patients with moderate mitral regurgitation but with otherwise favorable valve characteristics are not excluded from BMV. In our experience, the risk of resultant severe mitral regurgitation is minimal with the use of cautionary balloon sizing approach and the controlled stepwise dilation technique.^21-23 However, should the procedure fail to provide optimal clinical results, patients can still be subjected to elective surgery without exposing them to any additional risk. Recent work suggests that even patients with severe mitral regurgitation and mitral stenosis can benefit from BMV.²⁴

The presence of left atrial thrombus has traditionally been considered a contraindication to BMV because of the heightened risk of cardioembolism with the procedure. In patients with nonpedunculated thrombi in the left atrial cavity, one may elect to administer long-term (3-6 months) warfarin therapy (targeting an international normalized ratio between 2 and 2.5), if their clinical and hemodynamic status does not warrant immediate surgery and the mitral valves are deemed suitable for BMV. When the thrombus is observed to have resolved with transthoracic echocardiographic reassessments performed at 3-month intervals, BMV can then be performed safely after confirmation of the absence of thrombi in the left atrial cavity with transesophageal echocardiography.^23,25 Patients with lytic-resistant or mobile thrombi should be considered for open surgical commissurotomy with direct visual clot removal. Despite our safe and successful experience in performing Inoue-BMV in patients with left atrial thrombi confined to the appendage,^23,26 the subject has remained controversial even among experienced operators of Inoue-BMV. Therefore, the alternative approach is either to subject patients with appendage thrombi to mitral valve surgery or to defer BMV for stable patients until resolution of the thrombi under warfarin treatment.^23,27 Another clinical concern is pulmonary hypertension; successful BMV usually decreases the pulmonary pressure, although early intervention is better.²⁸

The best choice of technique for BMV remains a contentious issue. Beside the original Inoue technique using size-adjustable, self-positioning balloon catheters, various other techniques for performing BMV have been developed that use fixed-size balloon catheters. These include the antegrade (transvenous) approaches with 1 or 2 balloon catheters through 1 or 2 interatrial septal punctures^29,30 or the retrograde (transarterial) approaches with transseptal wiring or without transseptal access.³¹ An overview of the various comparative studies, both randomized and nonrandomized, does not reliably indicate that either the double-balloon or the Inoue-balloon technique is superior to the other in achieving a larger final mitral valve area. There is, however, persuasive evidence that the Inoue-balloon approach is technically less demanding and clearly simpler to perform; hence, it has a shorter irradiation and procedural time and is safer than the double-balloon approach.^32-34 These advantages are vital in pregnant patients in whom the hazards of irradiation to the fetus are of paramount importance and for patients with pulmonary edema in whom swift and expeditious BMV is clearly desirable. Therefore, the transvenous Inoue-balloon approach has obtained its current position as the principal BMV technique.

Subsequent sections of the present chapter are drawn largely from our incremental experience in Inoue-BMV techniques and evolving technical refinements in BMV techniques. These have resulted in a nearly 100% technical success rate and a significant diminution in complications despite the presence of a significant number of technically demanding scenarios and high-risk comorbid conditions.^12,22,23 Mainly discussed are the technical aspects of Inoue-BMV, the pitfalls and tricks to facilitate a successful procedure, and how to minimize procedure-related complications.

Transseptal catheterization is a vital component of BMV. Transseptal puncture must not only be executed safely to avoid cardiac perforation but also made at an appropriate interatrial septal site to facilitate balloon crossing of the stenosed mitral valve.³⁵ To avert cardiac perforation, some operators have resorted to routine intraprocedural transesophageal echocardiography to facilitate optimal transseptal needle placement; however, even with the echocardiographic guidance, cardiac perforation may still occur.³⁶ Recently developed intracardiac echocardiography is useful, but it is not widely available and its use adds to more cost. Therefore, acquisition of basic transseptal skill is essential. To perform transseptal procedure, biplane fluoroscopic equipment is preferable, but single-plane fluoroscopy is usually sufficient.

The instruments used for the procedure include a Brockenbrough needle and a 7- or 8-Fr transseptal catheter. The use of an outer sheath is optional, but its utility is recommended, especially for inexperienced operators, for 2 reasons: (1) to prevent inadvertent perforation of the dilator by the needle during its insertion, and (2) to prevent left atrial perforation during insertion of the catheter/needle into the left atrium; the sheath tip works as a safety stopper at the septum.

Landmarks for Optimal Puncture Site

The puncture target site is usually located at the cross point of (1) a horizontal line crossing the center of the mitral annulus (M-line) and (2) a vertical line, assumed to divide the interatrial septum into anterior and posterior halves (“midline”). However, in individual cases, the puncture site may have to be adjusted. For example, in patients with giant left atria, the operator is often forced to make the interatrial septal puncture more caudal to the horizontal M-line. In patients with a more vertically oriented left ventricle and a relatively small left atrium, the operator may need to make the puncture site slightly more cephalad to the M-line and lateral to the vertical midline.

Definition of Horizontal M-Line

The line is derived from a diastolic stop frame of diagnostic left ventriculography obtained in 30° right anterior oblique (RAO) view (Fig. 41-1). This horizontal line level is memorized in relation to the vertebral body. The angiogram is also used as a road map during transseptal puncture and balloon catheter manipulation.

Definition of Vertical Midline

Inoue devised a specific transseptal puncture technique designed for the Inoue balloon BMV, incorporating the concept of a vertical “midline,” a line assumed to divide the interatrial septum into anterior and posterior halves.³⁷ The midline is a vertical line crossing the midpoint of a horizontal line spanning the anterior and posterior limits of the septum. The upper end of the tricuspid is assumed to be the anterior septal limit in Inoue’s angiographic method, corresponding to the aortic valve in our modified fluoroscopic method.

Because the septum lies within the superimposed area between the 2 atria in both methods, the medial atrial silhouette (usually the left atrium) is used as the posterior limit (not necessarily the posterior border) because there is no septum beyond this silhouette. Infrequently, such as in patients with giant left atria or distorted cardiac anatomy, the right atrial border is medial to that of the left atrium, and thus, the right atrial border is used as the posterior limit.

1. 
   

   Angiographic method (Fig. 41-2A-B): The midline is defined based on the landmarks obtained from frontal plane right atrial angiography during normal respiration.

   
   
2. 
   

   Fluoroscopic method (Fig. 41-2C): In this method, the aortic valve is used to substitute the upper end of the tricuspid valve as the anterior septal limit because the 2 structures are in close proximity to each other. Therefore, a pigtail catheter is placed with its tip in the noncoronary sinus of Valsalva, touching the aortic valve. The midline thus derived is usually identical to that from Inoue’s angiographic method.

Because in most cases of mitral stenosis the left and right atrial silhouettes are visible under fluoroscopy (see Fig. 41-2C), the angiographic method is infrequently used and is reserved for the following situations: (1) for operators inexperienced with the transseptal puncture technique, (2) in cases in which atrial silhouettes are not well visualized under fluoroscopy, and (3) in extremely difficult cases of transseptal puncture (eg, in the presence of a giant left atrium³⁵ and kyphoscoliosis).³⁸ In these cases, it may be necessary to perform biplane (frontal and lateral) right angiography to properly visualize the atrial septal orientation and relative anatomic relationships of the atria, the tricuspid valve, and the aorta.

Confirmation of Optimal Puncture Site

Frontal View

Under frontal view, the needle-fitted transseptal catheter placed in the superior cava inserted from the right femoral vein is slowly withdrawn to align the catheter/needle on the midline. Reshaping of the distal Brockenbrough needle to make it more curved may be necessary to align the catheter tip with the midline. The needle tip should be constantly kept concealed slightly (2-3 mm) within the catheter tip before needle puncture.

The catheter tip should not be set medial to the midline to avoid puncturing the aorta, tricuspid valve, or coronary sinus. More importantly, the puncture site thus made is too close to the mitral valve, and this makes balloon crossing of the mitral valve difficult or even impossible. Slight lateral deviation of the puncture site to the midline is permissible, especially in patients with relatively small left atria. It is important to note that there may not be septum in an area near the inferior (caudal) border of the left atrium because the atrium often bulges caudally beyond the true septal boundary.

Thirty-Degree Right Anterior Oblique View

The catheter tip site is further examined in this view to confirm its optimal position (see Fig. 41-1B), also using the left ventriculogram as a road map (see Fig. 41-1A). The tip position is usually in front of the spinal column, and the distance between the intended puncture site and the mitral orifice (P to M in Fig. 41-1C) should exceed 1.3 times the vertebra width for easy catheter balloon crossing of the mitral valve. Because the catheter tip is away from the aorta, tricuspid valve, coronary sinus, and left atrial posterior border, needle puncture at this point pierces only the septum unless the puncture is made too caudally near the caudal edge of the left atrium in frontal view.

Confirmation of Catheter Tip at Septum

When setting of the catheter/needle at the septum is in doubt, septal flush or stain method may be used to examine the catheter/needle tip position. This is done by injecting a small amount of pure contrast medium contained in a 5-mL syringe attached to the proximal needle end.

Septal Flush Method

The contrast medium flushing outlines the right atrial margin of the septum,³⁵ and the optimal puncture site is at the curving segment of the septal outline (Fig. 41-3A, right panel).