Methods and subjects

We recruited all patients with isolated pulmonary valve stenosis who underwent balloon valvoplasties at a single tertiary centre, albeit by multiple operators, between December 1984 and August 1996. They included patients with an associated haemodynamically insignificant atrial septal defect or patent oval foramen. Only those patients who have been followed up for at least three years were enrolled. Their medical records and investigations were reviewed.

The technique used has been described in detail, and was similar to that performed by others.^{13, 14} The size of the balloon varied according to the measurement of the diameter of the abnormal pulmonary valve taken at their basal attachment of the hinge points of the leaflets, on echocardiography, and angiography when available. There was no uniform criterion for sizing the balloon amongst the different operators, although most used a ratio of 1.2 to 1, with the balloon being bigger than the valvar diameter. The gradient across the pulmonary valve, or the right ventricular pressures in those with critical pulmonary stenosis, were measured before and after the procedure. We used two balloons⁸ in 10 patients, in whom the diameter of the pulmonary valve ranged from 7 to 20 mm, with a median measurement of 12 mm. There was no consistent record of the number of inflations made in each procedure, but usually 2 or 3 inflations were made, nor was it noted if any of the balloons ruptured during the procedure. Different catheters were used over the period of time, including Omega, Berman, Accent, Meditech, Cook and Mansfield balloons. All procedures were carried out under general anaesthesia. The procedures were performed by five cardiologists over the study period, but all essentially used similar techniques and made comparable measurements.

Follow-up included clinical evaluation and echocardiographic studies, with cardiac catheterization being repeated only in exceptional circumstances. The maximum peak instantaneous gradient across the pulmonary valve, and the degree of pulmonary incompetence, were reviewed in all patients by cross-sectional echocardiography aided by pulsed and colour Doppler. Angiograms were available in 25 patients.

Since general anaesthesia can lower both the measured gradient and pressures in the right ventricle, unless they were unavailable, we used echocardiographic studies to quantitate the degree of valvar incompetence. The severity of the incompetence was then graded semi-quantitatively using the colour Doppler images obtained from cross-sectional echocardiography (Fig. 1). The incompetence was judged to be trivial when the diastolic flow reversed just below the valve, mild when reversal occurred midway between the valve and the bifurcation of the pulmonary trunk, moderate when reversal was seen at the level of the bifurcation, and severe when flow reversal occurred within the right and left pulmonary arteries.

Figure 1. We are assessing semi-quantitatively the degree of pulmonary incompetence using colour Doppler, utilizing the parasternal short axis view on cross sectional echocardiography. Figure 1a shows trivial incompetence, with diastolic reversal of flow seen for a short distance below the pulmonary valve. RA: right atrium; RVOT: right ventricular outflow tract; PT: pulmonary trunk; AO: aorta; LA: left atrium. Mild incompetence is shown in Figure 1b, with reversal occurring midway between the pulmonary valve and the bifurcation of pulmonary trunk. Moderate incompetence is shown in Figure 1c, with the reversal of flow seen at the bifurcation of the pulmonary trunk. Severe incompetence, with reversal of flow occurring within the right and left pulmonary arteries, is shown in Figure 1d. RPA: right pulmonary artery; LPA: left pulmonary artery.

Other criterions, such as right ventricular size, the extent of the regurgitant jet in relation to the diameter of the pulmonary valve, and its extent in the right ventricular outflow tract, were noted if available, though the analysis only incorporated the colour Doppler findings, these being available in all subjects.

Statistical methods

Comparisons between the two groups were performed using Student's two-tailed paired t-test. The degree of pulmonary valvar insufficiency, and the age at the procedure, were compared between the groups of patients using an unpaired t-test. The analysis of variance was used to study the relationship between the ratio of the size of the balloon to the diameter of the valve, and the severity of pulmonary valvar insufficiency. A p value of less than 0.05 was considered statistically significant.

Results

The median age at initial valvoplasty was 1.08 years, including 7 neonates (Table 1). Of the 47 patients, seven required more than one valvoplasty. They were analysed separately. The median duration of follow-up was 5.64 years, with a range from 3.00 to 14.26 years. We followed up 11 patients for more than 10 years.

Table 1. The age of patients at the initial pulmonary valvoplasty.

In all, 57 valvoplasties had been performed in 49 patients between December 1984 and August 1996. Nearly all patients had a significant gradient across the pulmonary valve, with a median gradient of 60 mmHg, and a range from 11 to 140 mmHg. Critical stenosis had been diagnosed in 4 neonates requiring an infusion of prostaglandin, which would have maintained the pulmonary arterial pressure, thereby reducing the gradient (Table 2).

Table 2. Comparison of variables between those patients who underwent a single valvoplasty as compared with those that required a second procedure.

The pulmonary valves were described as “thickened and doming” in 36 patients, and as relatively normal in 13 patients, the information being derived mainly from the echocardiograms. In 11 patients, pulmonary incompetence was noted prior to dilation, trivial in 10 patients, but moderate in one.

Pulmonary incompetence

Following valvoplasty, 26 of 42 patients (60%), who had one procedure demonstrated trivial or mild incompetence as shown by cross-sectional echocardiography and colour Doppler performed one day after the procedure (Table 3 and Fig. 2). Moderate incompetence was seen in three patients (7%) after the procedure, these patients having had no incompetence noted before. In five patients (12%), there was no incompetence noted immediately after the procedure, while the presence of immediate incompetence after the dilation was not recorded in 8 patients (20%). The one subject with moderate incompetence noted before the procedure suffered no change in the severity of valvar incompetence after. At follow-up, there was an increase from 3 to 15 in the number of patients with moderate or severe pulmonary incompetence (Fig. 3). None of them developed significant right ventricular dilation or paradoxical septal motion, at least within the limitations of short-to-medium term follow-up, and may have reflected little change in ventricular compliance, which may precede true volume loading. The higher the initial grade of pulmonary incompetence after the valvoplasty, the more likely it was to progress with time. The patients with moderate or severe pulmonary incompetence found on follow-up were those who had the procedure done at a younger age (Fig. 4).

Table 3. The outcome for patients following a single pulmonary valvoplasty. The age at time of the procedure, and the duration of follow-up, are cited as median values, with the range in brackets. The change in the severity of pulmonary regurgitation relates to findings prior to valvoplasty and the latest follow-up. The severity of regurgitation continued to increase progressively after valvoplasty in 12 of the 42 patients (see Fig. 3).

Figure 2. The change in degree of pulmonary incompetence after a single pulmonary valvoplasty.

Figure 3. The change in degree of pulmonary regurgitation on follow-up seen in those patients where such change occurred, and excluding the remaining 30 patients where there was no change.

Figure 4. The overall change in the grading of pulmonary regurgitation. Patients with severe pulmonary regurgitation were mainly babies who had required the procedure early in life. Patients with trivial regurgitation showed a wide range of distribution in age.

The majority, however (62%), had only a mild increase in the degree of pulmonary incompetence after a single valvoplasty over the period of intermediate to long-term follow-up (Table 3 and Fig. 2).

The 7 patients who required repeated valvoplasties had no or only trivial incompetence immediately following the second procedure. At the latest follow-up, however, three of these 7 patients were found to have moderate pulmonary incompetence.

In our patients, there was no significant difference in the initial pressure gradient across the pulmonary valve, nor in the ratio of the size of the balloon to the diameter of the pulmonary valve among the groups of patients, and the ratio did not significantly alter the outcome (Fig. 5).

Figure 5. The severity of pulmonary regurgitation after valvoplasty is compared with the ratio of the size of the balloon to the diameter of the pulmonary valve.

Discussion

We have charted the long-term results of relief of pulmonary valvar stenosis in 49 patients who underwent balloon valvoplasty. Our findings are generally comparable to previous reports, with several studies documenting the ongoing relief of obstruction at follow-up.^{6, 11, 12} These studies, for the most part, showed that the incompetence seen after the procedure was mild or haemodynamically insignificant. In part that may be related to reduced right ventricular compliance immediately after valvoplasty. In a series studying the results of percutaneous balloon valvoplasty in adults, it was reported that the mild incompetence noted immediately after the procedure may even resolve on follow-up.⁹

McCrindle and Kan¹¹ reported the results of a follow-up of about five years in 46 patients. They found a residual peak instantaneous Doppler gradient of 20 mmHg, with mild incompetence seen in four-fifths. Rao et al.⁶ reported the follow-up from 3 to 10 years in 80 children, with 51 having other associated cardiac lesions. They showed a residual peak instantaneous Doppler gradient of 17 mmHg. Pulmonary incompetence immediately after the procedure had increased in half, and was worse in four-fifths on long term follow-up, albeit that none of their patients was considered to have significant pulmonary incompetence.

In our study, 10 patients had trivial to mild incompetence prior to the procedure. It is noteworthy in this respect that trivial incompetence has been reported as a normal finding in healthy persons.¹⁵ The pulmonary valve coapts less well than the aortic valve, and the retrograde pressure gradient through the pulmonary valve in diastole may not be enough to close the valve as tightly as the aortic valve. Although four-fifths of our patients had an increase in the severity of pulmonary incompetence, only 7% showed moderate incompetence immediately after the procedure. In 13 patients (30%), we found a gradual increase in the severity of pulmonary incompetence at late follow-up, these findings being comparable to those reported by Rao et al.⁶ This funding may be due to changes in the pulmonary vascular resistance and right ventricular compliance with time.

Indeed, pulmonary incompetence has been found to be increased in most studies after balloon valvoplasty. It was increased in three-quarters of the patients studied by Ray et al.¹⁶ and in over seven-eighths of the cohort reported by McCrindle and Kan.¹¹ We should emphasize, nonetheless, that our system for grading pulmonary incompetence is only semi-quantitative. As stated by Feigenbaum,¹⁷ it is difficult to validate the degree of pulmonary regurgitation with Doppler echocardiography.^{15, 18, 19} William et al.²⁰ recently proposed more sophisticated methods for assessing pulmonary incompetence. Thus, angiograms can be considered as an adequate way for grading incompetence if the catheter can be placed in the distal pulmonary trunk, or one of its branches, with the catheter ports well away from the pulmonary valve. Serial noninvasive transthoracic echocardiographic studies, nonetheless, remain the mainstay for evaluation of these patients. Besides measuring the diastolic reversal in flow in the branches of the pulmonary trunk as we have described, others have used the ratio of the width of the regurgitant colour jet to the diameter of the pulmonary valve in early diastole as an objective method of assessing pulmonary incompetence severity using colour Doppler images.²⁰ Unfortunately clear documentation of this feature was not routinely available in our patients. Besides the regurgitant jet in the right ventricle, spectral Doppler deceleration, magnetic resonance imaging or even three-dimensional echocardiography, might also be used to assess the regurgitation and the right ventricular volume. But determination of right ventricular volumes is technically difficult because of right ventricular geometry. New technologies employing magnetic resonance imaging and assessing regurgitant fractions may provide the answer, though there is always the problem of studying an infant in a scanner. Prospective use of these techniques may, in future, provide a more accurate picture of outcome.

Be that as it may, we found that patients with more significant pulmonary incompetence on follow-up were those who had required the initial procedure at a younger age. That is probably because they had a more “abnormal” pulmonary valve, leading to significant stenosis and hence the need for earlier intervention. The period of follow-up was too short to determine the long-term impact of more severe pulmonary incompetence on right ventricular function, as is now known to occur in those adolescents and adults with moderate-to-severe pulmonary incompetence following surgical repair of tetralogy of Fallot.

In our study, we found no significant relationship between the grade of pulmonary incompetence and the ratio of the size of the balloon to the diameter of the pulmonary valve. The ratio varied from 0.65 to 2.14. Even in those in whom big balloons were used, there was no overt evidence of complications. There are reports, however, showing that larger balloons may cause damage to the right ventricular outflow tract,²¹ without conferring any additional benefits in relieving the valvar stenosis.¹⁵ Oversized balloons, with the ratio larger than 1.5, therefore, are not recommended. Those patients in whom the largest balloons were used relative to the diameter of the valve were those treated early in the series, when there was little information concerning the optimal size of balloon needed to relieve the stenosis.

In our study, therefore, we failed to demonstrate a significant relationship between the severity of the pulmonary incompetence as assessed by colour flow Doppler following the valvoplasty and the degree of incompetence noted before the procedure, nor the size of the balloon used relative to the diameter of the pulmonary valve. The severity of incompetence tended to increase with increasing years of follow-up. Age was the only significant factor in determining the severity of pulmonary incompetence noted after the valvoplasty in our patients. We postulated that those patients who required early intervention for pulmonary valvar stenosis tended to have more abnormal valves, and hence developed more severe pulmonary incompetence, after the percutanous balloon valvoplasty.