The incidence and risk factors for prosthetic pulmonary valve failure (PPVF) should be considered when determining optimal timing for pulmonary valve replacement (PVR) in asymptomatic patients with congenital heart disease (CHD). The cumulative freedom for reintervention due to PPVF after 146 PVR in 114 patients with CHD was analyzed. Six potential risk factors (underlying cardiac defect, history of palliative procedures, number of previous cardiac interventions, hemodynamic indication for PVR, type of intervention, and age at intervention) were analyzed using Cox proportional hazard modeling. Receiver operating characteristic (ROC) curves were used for discrimination. Internal validation in patients with tetralogy of Fallot was also performed. Median age at intervention was 23 years. There were 60 reinterventions due to PPVF (41%). Median event-free survival was 14 years (95% confidence interval [CI] 12 to 16 years). The only independent risk factor was the age at intervention (hazard ratio [HR] 0.93, 95% CI 0.90 to 0.97; p = 0.001; area under the ROC curve 0.95, 95% CI 0.92 to 0.98; p <0.001). The best cut-off point was 20.5 years. Freedom from reintervention for PPVF 15 years after surgery was 70% when it was performed at age >20.5 years compared with 33% when age at intervention was <20.5 years (p = 0.004). Internal validation in 102 PVR in patient cohort with tetralogy of Fallot (ROC area 0.98, 95% CI 0.96 to 1.0; p <0.001) was excellent. In conclusion, age at intervention is the main risk factor of reintervention for PPVF. The risk of reintervention is 2-fold when PVR is performed before the age of 20.5 years.
To relief right ventricular (RV) outflow tract obstruction in patients with tetralogy of Fallot and other congenital heart disease (CHD), it is often necessary to disrupt pulmonary valve integrity, which results in significant pulmonary regurgitation (PR) and RV volume overload. Chronic severe PR is well tolerated for years but eventually triggers a cascade of pathophysiological alterations, leading to RV dilation and dysfunction, atrial and ventricular arrhythmias, exercise intolerance, heart failure, and early mortality. Pulmonary valve replacement (PVR) usually results in disappearance or reduction of PR, improvement of functional class, and reduction of diastolic and systolic RV volumes. However, changes in RV systolic function, burden of arrhytmias, objective parameters of functional capacity, and improved survival have not been demonstrated. The PVR can be surgically or percutaneously performed, usually using a biologic valvular prosthesis. The mortality and morbidity of both procedures are low, but the functional integrity of all available tissue valves deteriorates with time leading to prosthetic pulmonary valve failure (PPVF) and need for reintervention. As the number of interventions is an independent predictor of death at long-term follow-up, knowledge of the risk for PPVF is crucial for optimal timing of PVR. However, the rate of prosthesis degeneration can be highly variable and the risk factors for PPVF are not well established. The main purpose of this study was to analyze the incidence and factors associated with PPVF in a large series of patients followed at a single center.
Methods
We conducted a retrospective cohort analysis of all patients who had undergone surgical PVR by a biologic prosthesis or valved conduit, treated in the adult CHD unit at La Paz University Hospital, from January 1990 to December 2013. Patients in which PVR was due to a non-CHD cause were excluded. All available medical records, including echocardiograms, magnetic resonance, or computed tomography imaging, hemodynamic studies and surgical protocols were revised. Follow-up was obtained from medical records supplemented of telephonic contact with patients themselves or their families. Complete follow-up was considered when a contact was available in the last year, a reintervention due to PPVF was performed, or the patient had died during follow-up. The study was approved by the local research ethic committee.
The PPVF was defined as the need for pulmonary valve reintervention, including either surgical or transcatheter PVR. Six potential risk factors for PPVF were determined after revision of the previous reports: (1) underlying cardiac defect; (2) previous palliative procedure; (3) number of previous intracardiac interventions; (4) hemodynamic cause of PVR; (5) age at intervention; and (6) type of PVR. The underlying cardiac defect was classified as (1) tetralogy of Fallot (including double outlet RV and pulmonary atresia with nonrestrictive ventricular septal defect); (2) pulmonary stenosis or atresia with intact ventricular septum; (3) transposition complexes (including complete transposition and congenitally corrected transposition); and (4) left ventricular outflow tract abnormalities repaired by Ross or Ross–Konno techniques. The hemodynamic cause of PVR was classified as predominant stenosis or regurgitation. The type of PVR was classified as porcine heterograft, pulmonary or aortic homograft, or valved conduit.
SPSS 15.0 for Windows (SPSS, Inc., Chicago, Illinois) was used for analysis. Quantitative values are presented as mean ± standard deviation or median and interquartile range when appropriate. Discrete data are presented as percentage of total number of patients or total number of pulmonary prostheses. The event-free time was analyzed by Kaplan–Meier method with 95% confidence intervals (CI), using as time scale the follow-up time between prosthetic valve insertion and date of reintervention or censoring. For determining risk factors of PPVF, a Cox proportional model was used, and hazard ratios (HR) with 95% CI were generated. As some patients had undergone 2 or more PVRs, the same patient could provide data for more than 1 time interval, and each new intervention was considered as an independent case. To analyze the effect of age at intervention on the incidence of PPVF, the series was subdivided by decades in 4 groups, and event-free times were compared using log-rank test. To discriminate the value of age at intervention on the incidence of PPVF, statistical C index was used, determining the area under the receiver operating characteristic (ROC) curve with 95% CI and chi-square test. The best cutpoint for age at intervention in the ROC curve was used for comparing actuarial curves of event-free time, using Kaplan–Meier method and the log-rank test. Internal validation in the tetralogy of Fallot cohort was also performed. A bilateral p value <0.05 was considered statistically significant in each analysis.
Results
The study included 146 interventions of PVR in 114 patients with CHD. Table 1 provides demographic data, underlying CHD, number of previous cardiac interventions, palliative procedures, hemodynamic indication for PVR, and type of intervention in the whole cohort. Median postoperative follow-up was 7 years (interquartile range 13 to 34). The follow-up was complete in 94% of cases. Twelve patients died (11.5%) during the adult follow-up. Patients who died had transposition complex more often as underlying CHD (p = 0.001) and more valved conduits (p = 0.008); however, there were no significant differences in age at PVR (p = 0.45), palliative procedures (p = 0.56), number of previous cardiac interventions (p = 0.18), or hemodynamic indication for PVR (p = 0.20). The cause of death was congestive heart failure in 4 cases, sudden death in 3, surgical reintervention in 3, and endocarditis in 2 cases. There were a total of 13 episodes of infective endocarditis after PVR in 13 patients.
| Number of pulmonary valve replacements | 146 |
| Male | 68 (60%) |
| Age at intervention (years) | 23 (13-34) |
| Age at end of the study | 36±11 |
| Underlying congenital heart disease | |
| Tetralogy of Fallot | 81 (71%) |
| Pulmonary stenosis or atresia | 12 (10.5%) |
| Transposition complexes | 12 (10.5%) |
| Ross/Ross-Konno procedures | 8 (7%) |
| Others | 1 (1%) |
| Prior palliative procedures | 49 (43%) |
| Number of previous cardiac interventions | 1 (1-2) |
| Hemodynamic indication | |
| Pulmonary valve stenosis | 81 (55%) |
| Pulmonary valve regurgitation | 57 (39%) |
| Others | 8 (6%) |
| Type of intervention | |
| Porcine heterograft | 85 (58%) |
| Valved conduits | 48 (33%) |
| Aortic or pulmonary homografts | 21 (14%) ∗ |
| Postoperative follow-up (years) | 7 (3-12) |
| Reintervention by prosthetic failure | 60 (41%) |
| Infective endocarditis | 13 (11.5%) |
| Death | 12 (10.5%) |
There were 60 cases of PPVF (41%) in 28 patients. Median event-free time was 14 years (95% CI 12 to 16 years). Cumulative survival free from PPVF at 10, 15, and 20 years since intervention was 71%, 41%, and 25%, respectively ( Figure 1 ). Table 2 demonstrates that there was no significant relation between time to PPVF and underlying CHD, previous palliative procedure, number of previous intracardiac interventions, hemodynamic indication for PVR, or type of PVR. The only independent risk factor for PPVF was age at intervention (p = 0.002). Figure 2 shows the event-free time curves in patients grouped by decade of age at intervention (p = 0.014 intergroups). The area under ROC curve of the interrelation between age at intervention and PPVF was 0.95 (CI 0.92 to 0.98; p <0.001). The cutpoint for age at intervention with the best discriminative value for PPVF was 20.5 years ( Figure 3 ). Freedom from reintervention due to PPVF 15 years after PVR was 70% when intervention was performed at age >20.5 years compared to 33% when PVR was undergone at age <20.5 years (p = 0.004; Figure 4 ). Internal validation in 102 PVR in the cohort with tetralogy of Fallot showed an area under ROC curve of 0.98 (CI 96.5 to 1.0; p <0.001; Figure 3 ). The best cutpoint was also 20.5 years, and freedom from reintervention due to PPVF 15 years after PVR was 64% when intervention was performed at age >20.5 years compared to 24% when PVR was undergone at age <20.5 years (p = 0.001) ( Figure 4 ).
| Hazard ratio | IC 95% | p | |
|---|---|---|---|
| Age at intervention (years) | 0.95 | 0.92-0.98 | 0.002 |
| Underlying heart defect | |||
| Tetralogy of Fallot | 0.83 | 0.28-2.1 | 0.51 |
| Pulmonary stenosis/atresia | 0.77 | 0.45-3.6 | 0.61 |
| Transposition complexes | 1.08 | 0.55-2.1 | 0.82 |
| Ross procedures | 1.44 | 0.52-4.0 | 0.48 |
| Hemodynamic indication | |||
| Stenosis | 1.47 | 0.79-2.7 | 0.22 |
| Regurgitation | 0.81 | 0.41-1.61 | 0.55 |
| Type of valve replacement | |||
| Porcine heterografts | 0.96 | 0.53-1.7 | 0.88 |
| Valved conduits | 1.10 | 0.65-1.9 | 0.73 |
| Homografts | 1.95 | 0.73-4.8 | 0.20 |
| Prior palliative procedures | 1.39 | 0.83-2.3 | 0.21 |
| Number of previous interventions | 1.13 | 0.80-1.6 | 0.49 |
