Abstract
Background:
Percutaneous balloon mitral valvuloplasty is the preferred therapeutic strategy in patients with mitral stenosis, but it has shortcomings in a subset of patients.
Hypothesis:
A new method of balloon sizing through echocardiographic measurement of the intercommissural diameter would be safe and effective and lead to better outcomes.
Methods:
Eighty‐six mitral‐stenosis patients were randomly assigned to undergo balloon mitral valvuloplasty either with height‐based balloon reference sizing (HBRS group, n = 43) or with balloons sized by the echocardiographic measurement of intercommissural diameter (EBRS group, n = 43). Postprocedural mitral valve area (MVA) and severity of mitral regurgitation (MR) were assessed via echocardiography and ventriculography. Intention‐to‐treat approach was applied for the statistical analysis.
Results:
Baseline characteristics were not different between the groups. The mean of the estimated balloon reference sizes was significantly higher in the HBRS patients than in the EBRS group (26.4 ± 0.92 mm, 95% confidence interval [CI]: 26.2–26.6 vs 24.5 ± 1.03 mm, 95% CI: 24.2–24.7, respectively; P = 0.006). Final MVAs were significantly larger in the EBRS group (1.5 ± 0.2 cm2, 95% CI: 1.46–1.59 vs 1.4 ± 0.2 cm2, 95% CI: 1.35–1.47, respectively; P = 0.01). The occurrence of new or aggravated MR was significantly lower in the EBRS group as assessed both by echocardiography (P = 0.04) and ventriculography (P = 0.05). Mitral regurgitation was aggravated in 13 (29.3%) patients in the HBRS group and in 5 (11.5%) patients in the EBRS group.
Conclusions:
Percutaneous balloon mitral valvuloplasty via the Inoue technique using balloons sized by the echocardiographic measurement of the maximal commissural diameter is an effective and safe method that might lead to an acceptable increase in the MVA and significant decrease in the rate and severity of iatrogenic MR.
The results of this work were orally presented in part at EuroPCR, Paris, France, May 17‐20, 2011. The authors have no funding, financial relationships, or conflicts of interest to disclose.
Rheumatic heart disease is no longer the most common cause of valvular heart disease in developed countries, but it is still affecting many areas of the world.1 Mitral stenosis is the most important long‐term sequel of rheumatic fever, and it has an unfavorable impact on the quality of life and survival of the patients involved. Gradual development of left‐heart obstruction gives rise to a variety of hemodynamic and structural changes that might be irreversible in patients with long‐standing disease. Drug therapy neither cures nor modifies the course of the disease, but it is routinely employed to reduce symptoms and prevent thromboembolic complications, or as prophylaxis against recurrent rheumatic fever.
Dedicated surgical and percutaneous procedures have revolutionized the management and outcome of these patients during recent decades. Percutaneous balloon mitral valvuloplasty (BMV) via the Inoue technique is the ideal method of treatment in patients with mitral stenosis and favorable valve characteristics and is preferred over surgery because of its lower morbidity and cost.2 On the other hand, the result of BMV in mitral‐stenosis patients with a high Wilkins score is less predictable and seems to be inferior to that in patients with pliable valves. Nevertheless, BMV may be indicated in some symptomatic patients who are at high risk for surgery, even when the valve morphology is not ideal.3., 4., 5. Successful BMV improves the hemodynamics and symptoms of mitral stenosis and has positive effects on the immediate‐ and long‐term survival of patients, with low rates of functional disability, need for surgery, or repeat BMV.6., 7., 8., 9., 10., 11.
Different studies have shown that acute procedural results, including final mitral valve area (MVA) and postprocedural mitral regurgitation (MR), independently predict the long‐term outcome after percutaneous BMV.12 Accurate adoption of a balloon catheter for the stepwise dilation protocol is a crucial stage during the procedure to reach an effective valve area without creating significant MR.13 Current guidelines rely on the balloon reference size, which is the expected maximal inflated balloon catheter diameter estimated with an empiric formula based on the height of the patient.14., 15., 16., 17. Thereafter, the balloon size has to be adjusted according to such echocardiographic and angiographic indices as pliability of the valve, calcification, and severity of MR to start the stepwise dilation technique. The safety and efficacy of the current standard balloon‐selection method is known and has been confirmed in many studies, but there are occasional cases that experience unexpected significant degrees of MR.
Given that height could not be considered as the only determinant of MVA, the effective (maximal) balloon dilating area derived from height might not necessarily apply to all patients with similar height and body surface area but with different body habitus, heart orientation, and configuration of the cardiac skeleton. We tested a more direct and conservative method of balloon sizing through the echocardiographic measurement of the maximal intercommissural distance of the mitral valve to avoid the possible over‐ and undersizing of the balloon and unfavorable results.
Methods
Study Design
This randomized controlled trial was a single‐center study on rheumatic mitral‐stenosis patients who were randomly allocated to undergo BMV either with conventional height‐based balloon reference sizing (HBRS group) or with balloons sized by the echocardiographic measurement of the diastolic intercommissural diameter (EBRS group) in a Zelen design.18 The local ethics committee of Rajaei Cardiovascular, Medical and Research Center, Tehran University of Medical Sciences, approved the trial design.
Patient Population and Randomization
Between April and October 2010, 142 patients with mitral stenosis were initially evaluated. Patients were considered eligible if they were age >18 years with symptomatic moderate to severe mitral stenosis (MVA <1.5 cm2). Asymptomatic patients also were eligible if they had pulmonary artery systolic pressure >50 mm Hg at rest or 60 mm Hg with exercise in the presence of favorable morphologic criteria of the mitral valve derived from the echocardiographic Wilkins scoring system.19
Exclusion criteria were mild mitral stenosis, Wilkins score >10, fluoroscopic mitral valve calcification, more than moderate (2+) MR, presence of left atrial clot, history of previous surgical or percutaneous commissurotomy, complications during the procedure except for MR, concomitant significant coronary artery or valvular disease requiring surgical correction, and current pregnancy.
Randomization was performed based on the computerized balanced block randomization method in blocks of 4: 43 patients to the HBRS group and 43 to the EBRS group. The patients who were allocated to the EBRS group (treatment group) were invited to participate and give their consent. Among them, 8 patients refused to take part and subsequently received HBRS. Finally, 51 patients received HBRS and 35 EBRS, but the analysis was performed with patients analyzed in the original randomized groups.
Procedural Protocol and Follow‐Up
The demographic and clinical data of the patients were recorded. All patients underwent a complete transthoracic and transesophageal 2‐dimensional Doppler and color flow study using the Vivid 3 cardiovascular ultrasound system (GE Medical Systems, Milwaukee, WI), and the following measurements were obtained: mitral valve mean and peak pressure gradients, severity of MR, pulmonary systolic pressure, right ventricular function, mitral valve Wilkins score, MVA by direct planimetry, and pressure half‐time.
Finally, 2 expert echocardiography fellows measured the maximal intercommissural distance on a parasternal short‐axis view from the anterolateral to posteromedial commissures in mid‐diastole. Severity of mitral regurgitation was evaluated using ≥2 echocardiographic methods. The presence of concomitant valvular heart disease and left atrial clot was also determined. Cardiac catheterization, including the left‐ and right‐heart hemodynamic study, was done and the severity of angiographic MR was determined via left ventriculography. Coronary angiography was conducted if it was clinically indicated.
In the HBRS group, balloons were selected using the height‐based reference size; in the EBRS group, balloons were selected according to the echocardiographic measurement. The formulas are detailed as follows. HBRS: calculation according to the standard height‐based formula (0.1 × height + 10), after rounding the patient's height (cm) to the nearest zero. EBRS: intercommissural distance in mid‐diastole rounded to the nearest zero.
The classic antegrade Inoue technique was utilized for BMV. After the atrial septostomy and appropriate septal dilation, 100 IU/kg of heparin was administered to achieve activating clotting time >250 seconds. The BMV procedures were done in a standard and similar fashion. Table 1 shows the mode of balloon‐size selection and stepwise dilation in the 2 groups. Echocardiographic and hemodynamic studies were repeated between the steps and at the end of the procedure. A final left ventriculogram was obtained to assess the severity of MR. Finally, MVA and MR severity were reassessed on the day after the procedure.
Table 1.
Mode of Balloon Catheter Selection
| HBRS Group | EBRS Group | |
|---|---|---|
| Balloon reference size | ||
| Wilkins score ≤8, MR ≤1+ | HBRS‐matched | EBRS‐matched |
| Wilkins score 9–10, 2+ MR | 1 size < HBRS‐matched | EBRS‐matched |
| Stepwise dilation procedure | ||
| Wilkins score ≤8, MR ≤1+ | Initial: HBRS −2 mm; increments: 1 mm, or 0.5 mm if unilateral commissural split, new or aggravated (1+) MR | Initial: EBRS −2 mm; increments: 1 mm, or 0.5 mm if unilateral commissural split, new or aggravated (1+) MR |
| Wilkins score 9–10, 2+ MR | Initial: HBRS −4 mm; increments: 1 mm (low pressure zone) or 0.5 mm (high pressure zone) | Initial: EBRS −2 mm; increments: 1 mm, or 0.5 mm if unilateral commissural split, new or aggravated (1+) MR |
Abbreviations: EBRS, echocardiographic balloon reference sizing; HBRS, height‐based balloon reference sizing; MR, mitral regurgitation.
Primary and Secondary Endpoints
The primary endpoint of this study was the severity of MR or the degree of MR changes after the procedure, and the secondary endpoint was postprocedural MVA (cm2). Successful BMV was defined as the final MVA ≥1.5 cm2 or at least a 50% increase in MVA, and no more than a 1+ increase in MR severity.
Statistical Analysis
The data are described as mean ± standard deviation for the interval and number (%) for the categorical variables. One sample Kolmogorov‐Smirnov test was used to investigate the fitness of distribution of the interval variables to Gaussian distribution. Intention‐to‐treat approach was applied for the statistical analyses.
Comparison of the endpoints between the study groups was performed using the Student t test for the interval, the Mann‐Whitney U test for the ordinal, and the Pearson χ 2 or Fisher exact tests for the nominal variables. A P value <0.05 was considered statistically significant. The statistical analyses were performed using SPSS 15 for Windows (SPSS Inc., Chicago, IL).
Results
Patient and Procedural Characteristics
Of the 142 patients who were initially evaluated, 86 patients were subjected to random assignment: 43 to standard BMV (HBRS) and 43 to BMV with EBRS (Figure 1). The study patients were comparable in terms of clinical and baseline echocardiographic and hemodynamic characteristics (Table 2). The mean age of the patients was 43.6 years (range, 20–63 y) and 87.2% were female. The mean height of the patients was comparable between the 2 groups (P = 0.12).
Figure 1.
Flow of study participants. Abbreviations: EBRS, echocardiographic balloon reference sizing; HBRS, height‐based balloon reference sizing.
Table 2.
Background and Preprocedural Data of the Study Participants
| HBRS [n = 43] | EBRS [n = 43] | P Value | |
|---|---|---|---|
| Age, y | 45 ± 10.1 | 42 ± 12.3 | 0.17 |
| Sex, n (%) | |||
| M: 11 (12.8) | 6 (14) | 5 (11.6) | |
| F: 75 (87.2) | 37 (86) | 38 (88.4) | 0.75 |
| Weight, kg | 65 ± 11.4 | 66 ± 10.3 | 0.62 |
| Height, cm | 160 ± 10.1 | 163 ± 8.2 | 0.12 |
| MVA, cm2 | 0.88 ± 0.19 | 0.90 ± 0.20 | 0.6 |
| Mitral mean gradient (mm Hg) | |||
| Echocardiography | 9.5 ± 4.5 | 9.8 ± 5.1 | 0.79 |
| Catheterization | 10.7 ± 5.9 | 13.8 ± 8.1 | 0.08 |
| Mitral valve score | 8.9 ± 0.89 | 8.7 ± 1.01 | 0.5 |
| PA pressure (mm Hg) | 48 ± 15.6 | 52 ± 21.9 | 0.26 |
| RV function, n (%) | |||
| Mild dysfunction | 9 (20.9) | 4 (9.3) | 0.06 |
| Mild to moderate dysfunction | 23 (53.5) | 21 (48.8) | |
| Moderate dysfunction | 8 (18.6) | 6 (14) | |
| Moderate to severe dysfunction | 3 (7) | 12 (27.9) | |
| MR severity, n (%) | |||
| Echocardiography | |||
| None | 11 (25.6) | 14 (32.5) | 0.11 |
| Mild | 21 (48.8) | 24 (55.8) | |
| Mild to Moderate | 11 (25.6) | 4 (9.3) | |
| Moderate | 0 (0) | 1 (2.3) | |
| Catheterization | |||
| None | 17 (39.5) | 11 (25.5) | 0.26 |
| 1+ | 26 (60.4) | 29 (67.4) | |
| 2+ | 0 (0) | 2 (6.9) | |
Abbreviations: EBRS, echocardiographic balloon reference sizing; F, female; HBRS, height‐based balloon reference sizing; M, male; MR, mitral regurgitation; MVA, mitral valve area; PA, pulmonary artery; RV, right ventricle.
There was no significant difference between the 2 groups in terms of weight (P = 0.62); it could, therefore, be assumed that the patients had similar body surface areas. Importantly, there were no significant differences in MVA and echocardiographic mitral valve scores between the 2 groups (P = 0.60 and P = 0.50, respectively). The majority of the patients had severe mitral stenosis with mean MVA about 0.89 cm2 (range, 0.5–1.4 cm2). The degree of MR was also comparable between the 2 groups before the procedure as measured both by echocardiography (P = 0.11) and left ventriculography (P = 0.26). It seems that the RV function was more deteriorated in the EBRS group (P = 0.06), although there was no difference in pulmonary artery pressures.
Postprocedural Characteristics
BMV was assessed as successful in 39 (90.7%) patients in the HRBS group and in 38 (88.3%) patients in the EBRS group (P = 0.41). The mean of the estimated balloon reference sizes was significantly higher in the HBRS patients than in the EBRS group (26.4 ± 0.92 mm, 95% confidence interval [CI]: 26.2–26.6 vs 24.5 ± 1.03 mm, 95% CI: 24.2–24.7, respectively; P = 0.006). The final balloon sizes (the size at which the procedure was stopped because of successful BMV, appearance, or aggravation of > + MR) were not significantly different between the groups (25.7 ± 1.1 mm, 95% CI: 25.4–26 vs 25.4 mm ± 1.4 mm, 95% CI: 25–25.8, respectively; P = 0.27).
Postprocedural data of the patients are presented in Table 3. Although there were no significant differences in postprocedural mean gradient and pulmonary pressure between the groups, the EBRS patients obtained significantly larger MVAs (1.5 ± 0.2 cm2, 95% CI: 1.46–1.59 vs 1.4 ± 0.2 cm2, 95% CI: 1.35–1.47, respectively; P = 0.01). Of note, despite the fact that the incidence of significant MR (3+ or 4+) was comparably low, the degree of MR changes was significantly greater in the HBRS group than in the EBRS group (P = 0.04 by echocardiography as is demonstrated in Figure 2, P = 0.05 by ventriculography). When echocardiography was considered as the reference, 88.4% of the EBRS patients did not have a change in MR severity compared with 69.8% of the patients in the HBRS group (P = 0.04). A larger number of patients in the HBRS group experienced 1+ increase in MR severity compared with the EBRS patients (20.9% vs 2.3%). The occurrence of MR change ≥2+ was comparable but infrequent in both groups.
Table 3.
Postprocedural Data of the Study Patients
| HBRS [n = 43] | EBRS [n = 43] | P Value | |
|---|---|---|---|
| Estimated reference balloon size | |||
| HBRS (mm) | 26.4 ± 0.92 | NA | 0.006 |
| EBRS (mm) | NA | 24.5 ± 1.03 | |
| Final balloon size | 25.7 ±1.1 | 25.4 ±1.4 | 0.27 |
| MVA (cm2) | 1.4 ± 0.2 | 1.5 ± 0.2 | 0.01 |
| Mitral mean gradient (mm Hg) | |||
| Echocardiography | 3.0 ± 0.95 | 2.7 ± 0.86 | 0.09 |
| Catheterization | 0.98 ± 1.86 | 1 ± 2.03 | 0.92 |
| PA pressure (mm Hg) | 22 ± 6.8 | 25 ± 13.0 | 0.14 |
| MR severity, n (%) | |||
| Echocardiography | |||
| None | 8 (18.6) | 12 (27.9) | 0.07 |
| Mild | 18 (41.9) | 22 (51.1) | |
| Mild to Moderate | 10 (23.3) | 4 (9.3) | |
| Moderate | 6 (14) | 4 (9.3) | |
| Severe | 1 (2.4) | 1 (2.3) | |
| Catheterization | |||
| None | 13 (30.2) | 9 (20.9) | 0.69 |
| 1+ | 21 (48.8) | 25 (58.13) | |
| 2+ | 8 (18.6) | 8 (18.6) | |
| 3+ | 1 (2.4) | 0 (0) | |
| 4+ | 0 (0) | 1 (2.3) | |
| MR change, n (%) | |||
| Echocardiography | |||
| No change | 30 (69.8) | 38 (88.4) | 0.04 |
| 1+ | 9 (20.9) | 1 (2.3) | |
| 2+ | 4 (9.3) | 3 (6.9) | |
| 3 and 4+ | 0 (0) | 1 (2.3) | |
| Catheterization | |||
| No change | 29 (67.4) | 37 (86.4) | 0.05 |
| 1+ | 13 (30.2) | 5 (11.6) | |
| 2+ | 1 (2.3) | 0 (0) | |
| 3 and 4+ | 0 (0) | 1 (2.3) | |
Abbreviations: EBRS, echocardiographic balloon reference sizing; HBRS, height‐based balloon reference sizing; MR, mitral regurgitation; MVA, mitral valve area; NA, not applicable; PA, pulmonary artery.
Figure 2.
Changes in the severity of mitral regurgitation after the procedure.
Discussion
Inoue‐balloon mitral commissurotomy is the procedure of choice in many patients with rheumatic mitral stenosis. Accurate balloon catheter sizing and a targeted stepwise dilation procedure are of paramount importance for a successful BMV. Routine balloon sizing based on the height of the patient has long been validated in many studies; be that as it may, it seems to be empiric without considering actual differences in size and orientation of the heart and its fibrous skeleton and, as a result, the area of the mitral valve annulus and orifice. This discrepancy could be quite considerable in some patients and result in the dilation of a balloon with an inappropriate dilating area within the mitral valve. This could lead to occasional unsatisfactory results in patients with good morphologic valve features and despite the performance of seemingly perfect procedures by trained operators. Adoption of a more direct method to overcome possible inefficiency of height as the sole indicator of the balloon size might be a solution.
When echocardiography is used for balloon sizing, the balloon reference sizes are smaller than those obtained with the usual method.20 Consequently, the dilation protocol is started and continued with smaller balloon sizes. Impressively, our study revealed that this technique might be associated with acceptable postprocedural MVAs and significantly lesser degrees of MR changes. Echocardiography‐derived balloon sizing in this study was associated with lesser degrees of MR changes; these changes were, however, mostly within the range of 1+ to 2+. Considering the fact that severe MR is also infrequent with the routine method, the long‐term clinical importance of such small reductions in the severity of MR should be addressed in the future. Be that as it may, avoiding severe MR, which has an important impact on the prognosis of patients, might be possible via this method. Patients who are tall, short, or asthenic, or patients in whom the balloon sizes determined by echocardiography and height are significantly disproportionate, are those who most benefit from this method.
It should be considered that the patients assigned to the present study had relatively good morphologic valve features and a more conservative BMV might even have better results in mitral stenosis patients with high mitral valve scores. On the other hand, accurate echocardiographic measurement of the commissural distance might be problematic in patients with severe subvalvular involvement.
Study Limitations
First and foremost among the limitations of the present study is that it is not sufficiently powered because of the small number of participants. Also, the patients in the HBRS group were not informed about their participation in the trial, which has ethical implications. In addition, the single randomized Zelen design is less efficient statistically than a clear‐cut 2‐group design and the true treatment effect is diluted because some subjects were allowed to decline the allocated treatment.21
Conclusion
Echocardiographic balloon sizing for BMV is a reasonable method with acceptable final MVAs and might reduce the risk of significant MR, especially in patients with disproportional balloon sizes obtained from the 2 methods.
Acknowledgements
The authors would like to thank Dr Jalal Norouzi, Dr Majid Kiavar, and Dr Saeed Alipourparsa for their great assistance and help in this study.
References
- 1. Iung B, Vahanian A. Epidemiology of valvular heart disease in the adult. Nat Rev Cardiol. 2011;8:162–172. [DOI] [PubMed] [Google Scholar]
- 2. Rahimtoola SH, Durairaj A, Mehra A, et al. Current evaluation and management of patients with mitral stenosis. Circulation. 2002;106:1183–1188. [DOI] [PubMed] [Google Scholar]
- 3. Yoshida Y, Kubo S, Tamaki S, et al. Percutaneous transvenous mitral commissurotomy for mitral stenosis patients with markedly severe mitral valve deformity: immediate results and long‐term clinical outcome. Am J Cardiol. 1995;76:406–408. [DOI] [PubMed] [Google Scholar]
- 4. Hung JS, Lau KW. Percutaneous transvenous mitral commissurotomy is an acceptable therapeutic alternative in patients with calcified mitral valve. J Invasive Cardiol. 1999;11: 362–363. [PubMed] [Google Scholar]
- 5. Wahl A, Meier B. Percutaneous mitral balloon valvuloplasty in nonideal patients: go for it without expecting too much. J Invasive Cardiol. 1999;11:359–361. [PubMed] [Google Scholar]
- 6. Hernandez R, Bañuelos C, Alfonso F, et al. Long‐term clinical and echocardiographic follow‐up after percutaneous mitral valvuloplasty with the Inoue balloon. Circulation. 1999;99: 1580–1586. [DOI] [PubMed] [Google Scholar]
- 7. Fawzy ME, Fadel B, Al‐Sergani H, et al. Long‐term results (up to 16.5 years) of mitral balloon valvuloplasty in a series of 518 patients and predictors of long‐term outcome. J Interv Cardiol. 2007;20:66–72. [DOI] [PubMed] [Google Scholar]
- 8. Fawzy ME. Long‐term results up to 19 years of mitral balloon valvuloplasty. Asian Cardiovasc Thorac Ann 2009;17: 627–633. [DOI] [PubMed] [Google Scholar]
- 9. Lung B, Vahanian A. Rheumatic mitral valve disease. In: Otto CM, Bonow RO, eds. Valvular Heart Disease: A Comparison to Braunwald' s Heart Disease. Philadelphia, PA: Saunders/Elsevier; 2007:481–501. [Google Scholar]
- 10. Carabello BA. Modern management of mitral stenosis. Circulation. 2005;112:432–437. [DOI] [PubMed] [Google Scholar]
- 11. Song JK, Song JM, Kang DH, et al. Restenosis and adverse clinical events after successful percutaneous mitral valvuloplasty: immediate post‐procedural mitral valve area as an important prognosticator. Eur Heart J. 2009;30:1254–1262. [DOI] [PubMed] [Google Scholar]
- 12. Jneid H, Cruz‐Gonzalez I, Sanchez‐Ledesma M, et al. Impact of pre‐ and postprocedural mitral regurgitation on outcomes after percutaneous mitral valvuloplasty for mitral stenosis. Am J Cardiol. 2009;104:1122–1127. [DOI] [PubMed] [Google Scholar]
- 13. Hung JS, Chern MS, Wu JJ, et al. Short‐ and long‐term results of catheter balloon percutaneous transvenous mitral commissurotomy. Am J Cardiol. 1991;67:854–862. [DOI] [PubMed] [Google Scholar]
- 14. Lau KW, Hung JS. A simple balloon‐sizing method in Inoue‐balloon percutaneous transvenous mitral commissurotomy. Cathet Cardiovasc Diagn. 1994;33:120–129. [DOI] [PubMed] [Google Scholar]
- 15. Lau KW, Hung JS. Balloon‐sizing method in Inoue mitral commissurotomy. Cathet Cardiovasc Diagn. 1995;35:183. [DOI] [PubMed] [Google Scholar]
- 16. Hung JS, Lau KW. Pitfalls and tips in Inoue‐balloon mitral commissurotomy. Cathet Cardiovasc Diagn. 1996;37: 188–199. [DOI] [PubMed] [Google Scholar]
- 17. Hung JS, Lau KW, Lo PH, et al. Complications of Inoue‐balloon commissurotomy: impact of operator experience and evolving technique. Am Heart J. 1999;138:114–121. [DOI] [PubMed] [Google Scholar]
- 18. Zelen M. A new design for randomized clinical trials. N Engl J Med. 1979;300:1242–1245. [DOI] [PubMed] [Google Scholar]
- 19. Wilkins GT, Weyman AE, Abascal VM, et al. Percutaneous balloon dilatation of the mitral valve: an analysis of echocardiographic variables related to outcome and mechanism of dilatation. Br Heart J. 1988;60:299–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Sanati HR, Kiavar M, Salehi N, et al. Percutaneous mitral valvuloplasty—a new method of balloon sizing based on commissural diameter to improve procedural results. Am Heart Hosp J. 2010;8:29–32. [DOI] [PubMed] [Google Scholar]
- 21. Torgerson DJ, Roland M. Understanding controlled trials: what is Zelen's design? BMJ. 1998;316:606. [DOI] [PMC free article] [PubMed] [Google Scholar]