The data describing the change in exercise capacity after surgical or interventional management of the patient with right ventricular (RV) outflow tract (OT) dysfunction are conflicting. The pathophysiologic consequences of RVOT interventions and the subsequent change in exercise performance are still poorly understood. We sought to assess the effect of percutaneous pulmonary valve implantation (PPVI) on exercise capacity in (1) patients with predominantly pulmonary stenosis (PS) and (2) in patients with predominantly pulmonary regurgitation (PR). A total of 63 patients with either predominantly PS (n = 37) or PR (n = 26) underwent PPVI. Cardiopulmonary exercise testing and magnetic resonance imaging were performed before and within 1 month after PPVI. On magnetic resonance imaging, the at rest effective biventricular stroke volumes improved in both groups after PPVI (p <0.001), but the ejection fraction improved only in the PS group. In the PS group, exercise capacity (peak oxygen uptake, p <0.001), ventilatory efficiency (p <0.001), and peak oxygen pulse (p <0.001) improved after PPVI. In the PR group, none of these parameters changed after PPVI (p = 0.6, p = 0.12, and p = 0.9, respectively). On multivariate analysis, the reduction in RVOT gradient was the only predictor of improved peak oxygen uptake when assessed in the whole patient group (r part = −0.59; p <0.001) or in the PS (r part = −0.45; p = 0.002) or PR groups alone (r part = −0.45; p = 0.02). In conclusion, acutely after PPVI, exercise capacity improves with the relief of stenosis but not regurgitation. A reduction in the RVOT gradient, even small gradients, was the only independent predictor of improved peak oxygen uptake in both patient groups, irrespective of improved pulmonary valve competence.
The data are conflicting describing the change in exercise capacity after surgical or interventional management of the patient with right ventricular (RV) outflow tract (OT) dysfunction. Furthermore, the issue is complicated because different physiologic lesions (stenotic or regurgitant) might respond differently to treatment when the exercise capacity is tested. A better understanding of the effect of altering different pathologic RV loading conditions (RV pressure overload vs RV volume overload) on the maximum and submaximum parameters of exercise capacity might help to refine the selection criteria and timing of RVOT interventions. In the present study, we assessed the effect of percutaneous pulmonary valve implantation (PPVI) on the maximum and submaximum parameters of exercise capacity in (1) patients with predominantly pulmonary stenosis (PS) and (2) in patients with predominantly pulmonary regurgitation (PR). Furthermore, we sought to identify the variables that would predict a change in exercise capacity after PPVI.
Methods
A total of 63 patients were included in the present prospective study. Patients were included if they had either a predominantly stenotic lesion before treatment (defined by a peak systolic gradient across the RVOT >50 mm Hg on the echocardiogram and <25% PR fraction on magnetic resonance imaging [PS group]); or a predominantly regurgitant lesion (defined by a peak systolic gradient across the RVOT <50 mm Hg on the echocardiogram and >25% PR fraction on magnetic resonance imaging [PR group]). The patients also had to fulfill the morphologic criteria for PPVI, as previously published. Those patients who presented with mixed RVOT lesions (who did not fulfill the criteria for either a predominantly stenotic lesion or a predominantly regurgitant lesion) were excluded from the present report. Other exclusion criteria were a contraindication for magnetic resonance imaging and a contraindication for exercise testing.
Valve implantation (Melody, Medtronic, Minneapolis, Minnesota) was performed with the patient under general anesthesia. Pressure measurements were performed before and immediately after PPVI, as previously reported. The patients were classified according to the New York Heart Association functional class before and after PPVI. All patients and parents provided written informed consent, as appropriate. The ethics committees at the 2 contributing institutions approved the study protocol.
Paired cardiopulmonary exercise testing and magnetic resonance imaging were performed during the same visit, before PPVI and a median of 4 days (range 3 to 42) after PPVI. Cardiopulmonary exercise testing was performed on a bicycle ergometer. After an unloaded warm-up, the workload was increased by 10 to 20 W/min, and the patients were encouraged to exercise until exhaustion after about 10 minutes of loaded exercise. The tests were considered maximal with a respiratory exchange ratio of ≥1.09. The 12-lead electrocardiogram was monitored continuously. The blood pressure was recorded every 2 minutes during exercise. Breath-by-breath respiratory gas exchange measurements were recorded throughout the test. The peak oxygen uptake was defined as the average of the values obtained in the last 20 seconds of exercise. The anaerobic threshold was determined by the modified V-slope method. Ventilatory efficiency, defined as the slope between minute ventilation and carbon dioxide elimination was obtained by linear regression analysis using breath-to-breath data acquired throughout the exercise period. In addition, the peak oxygen pulse (peak oxygen uptake/peak heart rate) was calculated.
Magnetic resonance imaging was performed at 1.5 T (Symphony Maestro Series and Avanto, Siemens Medical Solutions, Erlangen, Germany). Retrospective gated steady-state free-precession cine images of the heart and arterial flow data with a flow-sensitive gradient echo sequence were acquired. Assessment of the RV and left ventricular volumes was performed by manually defining the endocardial outline at end-diastole and end-systole in each of the short-axis cine images (Argus, Siemens Medical Systems). The end-diastolic volume and end-systolic volume were calculated with Simpson’s rule for each ventricle, and from these volumes, the stroke volume and ejection fraction were derived. The detailed imaging protocols for assessment of biventricular function and great vessel blood flow have been previously described. The PR fraction was calculated as the percentage of backward flow over forward flow. Where PR was present, an effective RV stroke volume was calculated to reflect the net forward blood flow into the pulmonary arteries as follows: effective RV stroke volume = total pulmonary artery forward flow − pulmonary artery backward flow. All volume and flow measurements were indexed to the body surface area (ml/m 2 and ml/m 2 /s).
Data are expressed as the mean ± SD. The proportions are expressed as percentages. Two paired samples were analyzed with the paired Student t test. The parameter values before PPVI were compared between the 2 groups using the unpaired Student t test or Mann-Whitney U test. Categorical variables were compared using Fisher’s exact test. A Wilcoxon signed ranks test was used to assess the change in New York Heart Association functional class before and after PPVI. Linear regression analysis was performed to study the relation between both changes in hemodynamic variables and in biventricular function and changes in peak oxygen uptake and ventilatory efficiency after PPVI. Univariate regression analysis was performed after verifying normal P–P plots of regression-standardized residuals. A backward multivariate linear regression analysis was performed (probability of F-to-remove was ≥0.10) using the independent variables with p values <0.05 on univariate analysis. Part (also known as semipartial) correlation coefficients (r part ) were used to describe the actual association. All statistical tests were 2-sided, and p <0.05 was considered statistically significant. Statistical testing and data analysis were performed using Statistical Package for Social Sciences, version 16.0, for Mac (SPSS, Chicago, Illinois).
Results
The patient characteristics are listed in Table 1 . The mean age at study onset was 22.2 ± 11.5 years; 46% were female. Most patients (60%) had tetralogy of Fallot or a variant of this morphology. Of the 63 patients, 56 (89%) had had a right ventricle to pulmonary artery conduit placed at a previous operation (homograft in 87% of cases). The baseline characteristics, including age, diagnosis, and RVOT morphology did not differ significantly between the PS and PR groups. Although not statistically significant, more males were in the PS group than in the PR group (65% vs 39%). The patients were slightly more symptomatic in the PS group (22% with functional New York Heart Association class III) than in the PR group (12%); this difference was not significant (p = 0.17). The 2 groups were significantly different in terms of echocardiographically determined RVOT gradient (77.8 ± 17.4 vs 33.1 ± 11.5 mm Hg; p <0.001) and RV systolic pressure (85.0 ± 17.4 vs 46.0 ± 13.2 mm Hg; p <0.001).
| Parameter | Total Population (n = 63) | PS Group (n = 37) | PR Group (n = 26) | p Value (PS vs PR) |
|---|---|---|---|---|
| Mean age (years) at implant | 22 ± 12 | 23 ± 12 | 22 ± 11 | 0.41 |
| Female | 29 (46%) | 13 (35%) | 16 (61%) | 0.05 |
| Tetralogy of Fallot variant | 38 (60%) | 21 (57%) | 17 (65%) | 0.6 |
| Pulmonary stenosis | 13 | 5 | 8 | |
| Pulmonary atresia | 21 | 13 | 8 | |
| Absent pulmonary valve | 4 | 3 | 1 | |
| Double outlet right ventricle | 4 (6%) | 2 (5%) | 2 (8%) | 1 |
| After Rastelli operation | 7 (11%) | 6 (16%) | 1 (4%) | 0.22 |
| Ross procedure | 6 (10%) | 4 (11%) | 2 (8%) | 1 |
| Truncus arteriosus | 5 (8%) | 2 (5%) | 3 (12%) | 0.64 |
| Other | 3 (5%) | 2 (5%) | 1 (4%) | 1 |
| Right ventricular outflow tract characteristics | ||||
| Homograft | 55 (87%) | 35 (95%) | 20 (77%) | 0.06 |
| Hancock | 1 (2%) | 1 (3%) | 0 | 1 |
| Native/patch-extended outflow tract | 6 (10%) | 1 (2%) | 5 (19%) | 0.07 |
| Other | 1 (2%) | 0 | 1 (4%) | 0.41 |
| No. of heart operations | ||||
| Open | 2 (1–4) | 2 (1–4) | 1 (1–4) | 0.21 |
| Closed | 1 (0–2) | 1 (0–2) | 0 (0–2) | 0.55 |
| Total | 2 (1–5) | 3 (1–5) | 2 (1–5) | 0.19 |
| New York Heart Association class | ||||
| 1 | 10 (16%) | 7 (19%) | 3 (12%) | 0.5 |
| 2 | 42 (67%) | 22 (60%) | 20 (77%) | 0.17 |
| 3 | 11 (17%) | 8 (22%) | 3 (12%) | 0.33 |
| 4 | 0 | 0 | 0 | |
| Echocardiography | ||||
| Right ventricular systolic pressure (mm Hg) (n = 56) | 69 ± 25 | 85 ± 17 | 46 ± 13 | <0.001 |
| Peak right ventricular outflow tract gradient (mm Hg) | 59 ± 29 | 78 ± 17 | 33 ± 12 | <0.001 |
PPVI led to a significant improvement in the New York Heart Association functional class in both groups (6 of 37 patients with New York Heart Association class I before vs 26 of 37 after PPVI in the PS group and 3 of 26 vs 17 of 26 in the PR group; p <0.001). The invasive pressure measurements are summarized in Table 2 . According to the inclusion criteria and grouping of patients, patients in the PS group had a significantly greater RV systolic pressure (70.6 ± 16.0 vs 48.5 ± 13.5 mm Hg; p <0.001) and RVOT gradient (48.1 ± 18.7 vs 18.3 ± 11.9 mm Hg; p <0.001) before PPVI than did the patients in the PR group. PPVI resulted in a significant reduction in RV systolic pressure and pulmonary artery to RV pullback gradient in both groups. Although the end-diastolic pressure decreased in both groups, the change was significant only in the PS group.
| Variable | PS Group (n = 37) | PR Group (n = 26) | ||||
|---|---|---|---|---|---|---|
| Before | After | p Value | Before | After | p value | |
| Invasive hemodynamics | ||||||
| Right ventricular systolic pressure (mm Hg) | 71 ± 16 ⁎ | 44 ± 9 | <0.001 | 49 ± 14 | 37 ± 11 | <0.001 |
| Right ventricular end-diastolic pressure (mm Hg) | 12 ± 4 | 10 ± 4 | 0.001 | 11 ± 3 | 10 ± 5 | 0.18 |
| Right ventricular outflow tract gradient (mm Hg) | 49 ± 19 ⁎ | 20 ± 10 | <0.001 | 19 ± 12 | 12 ± 9 | 0.001 |
| Aortic systolic pressure (mm Hg) | 91 ± 14 | 102 ± 12 | <0.001 | 90 ± 11 | 97 ± 17 | 0.031 |
| Aortic diastolic pressure (mm Hg) | 54 ± 10 | 60 ± 9 | 0.003 | 52 ± 8 | 56 ± 9 | 0.09 |
| Magnetic resonance imaging | ||||||
| Right ventricular end-diastolic volume (ml/m 2 ) | 96 ± 29 ⁎ | 88 ± 24 | <0.001 | 117 ± 35 | 95 ± 33 | <0.001 |
| Right ventricular end-systolic volume (ml/m 2 ) | 51 ± 27 | 40 ± 21 | <0.001 | 55 ± 27 | 49 ± 28 | 0.004 |
| Right ventricular stroke volume (ml/m 2 ) | 45 ± 12 ⁎ | 48 ± 8 | 0.09 | 61 ± 12 | 46 ± 9 | <0.001 |
| Effective right ventricular stroke volume (ml/m 2 ) | 41 ± 11 | 47 ± 8 | 0.001 | 37 ± 6 | 45 ± 8 | <0.001 |
| Right ventricular ejection fraction (%) | 49 ± 15 | 56 ± 13 | <0.001 | 55 ± 10 | 52 ± 11 | 0.1 |
| Pulmonary regurgitation fraction (%) | 8 ± 7 ⁎ | 1 ± 3 | <0.001 | 39 ± 9 | 3 ± 5 | <0.001 |
| Left ventricular end-diastolic volume (ml/m 2 ) | 70 ± 14 | 77 ± 15 | 0.002 | 67 ± 15 | 73 ± 13 | 0.003 |
| Left ventricular end-systolic volume (ml/m 2 ) | 28 ± 10 | 28 ± 9 | 0.6 | 25 ± 8 | 26 ± 7 | 0.4 |
| Effective left ventricular stroke volume (ml/m 2 ) | 41 ± 9 | 47 ± 8 | <0.001 | 40 ± 7 | 46 ± 6 | <0.001 |
| Left ventricular ejection fraction (%) | 61 ± 9 | 64 ± 8 | 0.02 | 63 ± 6 | 65 ± 5 | 0.09 |
| Cardiopulmonary exercise testing | ||||||
| Predicted peak oxygen uptake (%) | 60 ± 13 | 68 ± 14 | <0.001 | 66 ± 17 | 67 ± 18 | 0.26 |
| Peak oxygen uptake (ml/kg/min) | 23 ± 7 | 26 ± 7 | <0.001 | 25 ± 8 | 25 ± 7 | 0.6 |
| Oxygen uptake at anaerobic threshold (ml/kg/min) | 14 ± 4 | 16 ± 4 | 0.025 | 15 ± 5 | 14 ± 5 | 0.6 |
| Peak work (Watts) | 124 ± 39 ⁎ | 134 ± 45 | 0.06 | 102 ± 39 | 105 ± 38 | 0.09 |
| Peak oxygen pulse (ml/beat) | 9 ± 2 | 10 ± 3 | <0.001 | 8 ± 3 | 8 ± 2 | 0.9 |
| Ventilatory efficiency slope | 35 ± 8 | 30 ± 4 | <0.001 | 34 ± 6 | 32.2 ± 7 | 0.12 |
| Heart rate at rest (beats/min) | 84 ± 16 | 82 ± 12 | 0.51 | 84 ± 11 | 82 ± 13 | 0.37 |
| Heart rate at peak exercise (beats/min) | 164 ± 16 ⁎ | 161 ± 16 | 0.23 | 154 ± 20 | 157 ± 21 | 0.36 |
| Predicted heart rate at peak exercise (%) | 83 ± 8 ⁎ | 82 ± 10 | 0.3 | 78 ± 9 | 79 ± 10 | 0.36 |
| Respiratory exchange ratio at peak exercise | 1.15 ± 0.04 | 1.15 ± 0.05 | 0.97 | 1.13 ± 0.06 | 1.13 ± 0.06 | 0.45 |
⁎ Significant difference (p <0.05) before PPVI between PS and PR groups.
Biventricular volumetric assessment and calculation of the great vessel blood flow showed an improvement in effective RV and left ventricular stroke volume at rest in both groups ( Table 2 ). In the PS group, this was a result of a decreased RV end-systolic volume and hence improved RV ejection fraction. In contrast, the RV ejection fraction remained unchanged in the PR group. In the latter patient group, the improvement in RV and left ventricular effective stroke volume resulted from abolished PR. Although relief of the RV volume overload resulted in a reduction in the RV end-diastolic volume and RV end-systolic volume, the total RV stroke volume decreased in these patients with predominantly PR. In both groups, PPVI led to a greater left ventricular end-diastolic volume, indicating improved left ventricular filling, and to an improvement in left ventricular ejection fraction, which was only significant in the PS group.
The results of the cardiopulmonary exercise testing are listed in Table 2 . Before the procedure, the peak workload and percentage of predicted peak heart rate were significantly lower in the PR group. No difference was noted between the 2 groups in terms of exercise capacity, ventilatory efficiency, or exercise level. After PPVI, significant improvement occurred in the absolute measures of peak oxygen uptake, percentage of predicted peak oxygen uptake, peak workload, oxygen uptake at anaerobic threshold, peak oxygen pulse, and slope between minute ventilation and carbon dioxide elimination in the PS group. In contrast, patients who were treated for predominantly PR showed no improvement in any of these parameters. The peak heart rate and respiratory exchange ratio did not change after PPVI in either group.
On univariate analysis, a greater reduction in the RV end-diastolic pressure, invasively measured RVOT gradient, and RV systolic pressure and an increase in the RV ejection fraction were all associated with improved peak oxygen uptake. With the exception of the change in RV end-diastolic pressure, the same parameters correlated significantly with the change in slope between minute ventilation and carbon dioxide elimination ( Table 3 ). In contrast, patients with a greater reduction in PR fraction were less likely to have improved peak oxygen uptake or slope between minute ventilation and carbon dioxide elimination. On multivariate analysis, a reduction in the RVOT gradient was the only predictor of improved peak oxygen uptake when assessed for the whole patient group or the PS group or PR group alone ( Figure 1 ). Similarly, a reduction in RV systolic pressure was the only significant predictor of improved minute ventilation/carbon dioxide elimination considering the whole study group or the PS or PR group separately.
