Surgical implantation of a right ventricle to pulmonary artery (RV-PA) conduit is an important component of congenital heart disease (CHD) surgery, but with limited durability, leading to re-intervention. The present single-center, retrospective, cohort study reports the results of surgically implanted RV-PA conduits in a consecutive series of children and adults with CHD. Patients with CHD referred for RV-PA conduit surgical implantation (from October 1997 to January 2022) were included. The primary outcome was conduit failure, defined as a peak gradient above 64 mm Hg, severe regurgitation, or the need for conduit-related interventions. Longitudinal echocardiographic studies were available for mixed-effects linear regression analysis. A total of 252 patients were initially included; 149 patients were eligible for follow-up data collection. After a median follow-up time of 49 months, the primary study end point occurred in 44 (29%) patients. A multivariable Cox regression model identified adult age (>18 years) at implantation and pulmonary homograft implantation as protective factors (hazard ratio 0.11, 95% confidence interval [CI] 0.02 to 0.47 and hazard ratio 0.34, 95% CI 0.16 to 0.74, respectively). Fever within 7 days of surgical conduit implantation was a risk factor for early (within 24 months) failure (odds ratio 4.29, 95% CI 1.41 to 13.01). Long-term use of oral anticoagulants was independently associated with slower progression of peak echocardiographic gradient across the conduits (mixed-effects linear regression p = 0.027). In patients with CHD, the rate of failure of surgically implanted RV-PA conduits is higher in children and after nonhomograft conduit implantation. Early fever after surgery is a strong risk factor for early failure. Long-term anticoagulation seems to exert a protective effect.
Surgical implantation of right ventricle to pulmonary artery (RV-PA) conduits is an important component of congenital heart disease (CHD) surgery either at the time of primary repair/palliation or later in life because of right ventricular outflow tract (RVOT) (or previous conduit) dysfunction. In the population of adults with CHD, it is estimated that ∼20% of patients have had RVOT revision at the time of the initial intracardiac repair, and at least a subset of these underwent an RV-PA conduit implantation along the therapeutic road.
A wide array of conduit types have been used in the last decades (including aortic/pulmonary homografts, composite grafts, and decellularized bioprosthetic conduits). Limited durability, increased risk of endocarditis, and need for surgical or transcatheter re-intervention have been reported in this patient population.
Multiple investigations have reported on the long-term results after RV-PA conduit implantation in patients with CHD. Reported risk factors for early failure include smaller conduit diameter, complex CHD with nonanatomical conduit placement, specific conduit types, younger age at conduit implantation, smoking, and higher body mass index. , , , , Mechanisms of conduit deterioration over time and results of different types of conduits across different age groups are key areas of interest.
Current evidence regarding the durability of RV-PA conduits largely concerns nonmodifiable risk factors for early conduit failure (such as age at intervention, specific type of CHD, conduit location). The present single-center, retrospective, cohort study reports on long-term performance after surgical implantation of RV-PA conduits in a consecutive series of patients with CHD treated using a uniform and standardized approach with different conduit types. The study also includes longitudinal echocardiographic serial assessment and data collection on 2 prespecified and potentially modifiable and patient-specific conduit failure risk factors/attenuators: postsurgical fever and long-term anticoagulation.
Methods
Patients with CHD referred to our institution for RV-PA conduit surgical implantation from October 1997 to January 2022 were included. A standardized, retrospective chart review was used to collect data regarding demographics, primary congenital anomaly, surgical history, co-morbidities, imaging data, index surgical intervention procedural data, periprocedural variables (including early postoperative fever, long-term anticoagulation, periprocedural mortality), longitudinal clinical variables, and relevant imaging data. In our institution, surgical patients are routinely followed up approximately 2 weeks after surgery. Subsequent follow-up occurs approximately every 9 to 12 months for the first 2 years, and every 2 years thereafter. A detailed list of variables collected for the study is reported in Supplementary Table 1 .
The primary study outcome was conduit failure, defined as the early occurrence of the postimplantation peak gradient above 64 mm Hg (4 m/s of jet velocity measured using continuous wave Doppler echocardiography) and/or severe “pulmonary/conduit” (see Supplementary Materials for details). , Review of imaging studies was performed by a single investigator (ECD) blinded to patient identifiers.
The secondary outcome was longitudinal change (increase) of continuous wave Doppler–derived peak gradient across the conduit in postoperative echocardiographic studies.
Exploratory outcomes included overall procedural mortality and periprocedural adverse events.
Details of standardized pulmonary homograft harvesting and preparation in our center are reported in the Supplementary Materials .
Between-group comparisons for clinical and outcome variables were performed using the independent samples t test, Wilcoxon rank-sum test, chi-square analysis, or Fisher’s exact test with appropriate variable-specific denominators.
The primary study end point was conduit failure as previously defined. Prespecified candidate predictors of conduit failure were (1) pulmonary homograft; (2) adult versus children (dichotomizing category) at the time of surgical implantation; (3) conduit size treated as a continuous variable and as a binary variable (>22 vs <22 mm); (4) indication for conduit implantation; (5) fever in the first 7 days after conduit implantation; and (6) long-term use of anticoagulation. Covariates for the conduit-related end point were chosen on the basis of extensive analysis of the existing research and included age at intervention, type of CHD, indication for intervention, type of conduit, conduit diameter, long-term use of anticoagulation, and fever within 7 days from surgery.
The time to primary outcome was computed using Kaplan–Meier estimates, and the log-rank test was used to compare survival distribution across prespecified binary groups. Cox modeling was used to compute the hazard ratio (with 95% confidence interval) of the primary end point. Details of the Cox regression model are reported in the Supplementary Materials .
Logistic regression was used to model the effect of postsurgical fever on early conduit failure (defined as conduit failure occurring before 24 months from the date of surgical implantation), and model robustness was assessed using bootstrapping resampling of the data set (500 replications). Both in the Cox regression model and the logistic regression model, the age at intervention was retained.
Mixed-effects linear regression model was used to test if long-term anticoagulation use was associated with slower longitudinal RV-PA conduit gradient increase over time. Details of this analysis are reported in the Supplementary Materials .
Data are reported as mean ± SD, median (first and third quartile), or frequency (%). All tests were 2-sided. Analysis was performed using STATA 17 data analysis software (StataCorp LLC, College Station, Texas).
The institutional review board of IRCCS Azienda Ospedaliera-Universitaria di Bologna approved the study and granted a waiver for patient consent given the retrospective nature of the study, with minimal patient interaction.
Results
Between October 18, 1997 and January 10, 2022, 252 patients with CHD underwent surgical implantation of an RV-PA conduit in our center. More than 95% of the procedures were accomplished by 3 experienced primary operators. Figure 1 summarizes the structure of this observational cohort as per the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) approach. The data lock was on December 31, 2022. Table 1 summarizes pertinent demographic, clinical, and procedural data for the entire cohort and by age at the time of the surgical procedure. As expected, the 2 groups of patients differed in a subset of important clinical and procedural variables, including body size, age, type of primary CHD, type of conduit, conduit diameter, and procedural data. Conotruncal abnormalities were the primary CHD for most patients. In 145 of 252 cases (∼58%), a previous surgical procedure occurred before the index procedure, and 71 patients (28%) already had an RV-PA conduit implanted (which was replaced at the time of the index procedure). The 3 most common indications for conduit placement were “pulmonary regurgitation” (which was the most common indication in the adult subgroup), primary CHD repair (which was the most common indication in the children subgroup), and RVOT obstruction. In 81 patients (32%), an additional surgical component was needed beyond conduit implantation. Pulmonary homografts were the most common types of conduits implanted both in the children and the adult subgroup; no aortic homograft was used in this study. Furthermore, no Ozaki reconstruction of the RVOT or MASA Valve implantation was performed in this patient cohort.
| Variable | Overall cohort (n=252) | Children (<18 years) (n=127) | Adults (> 18 years) (n=125) | p-value |
|---|---|---|---|---|
| Age at intervention, years | 16 (6-31) | 8 (1-14) | 33 (23-47) | <0.001 |
| Weight, Kg | 55 (29-70) | 29 (15-55) | 65 (55-75) | <0.001 |
| Height, cm | 160 (135-170) | 135 (97-162) | 166 (166-174) | <0.001 |
| BSA, m 2 | 1.01 (0-1.68) | 0.58 (0-1.33) | 1.56 (1.35-1.83) | <0.001 |
| Primary CHD, n (%) | <0.001 | |||
| Tetralogy of Fallot | 88 (35%) | 31 (25%) | 57 (46%) | |
| Tetralogy of Fallot with PA | 45 (18%) | 32 (25%) | 13 (10%) | |
| Truncus arteriosus | 29 (12%) | 28 (22%) | 1 (1%) | |
| Pulmonary stenosis | 25 (10%) | 6 (5%) | 19 (15%) | |
| TGA-related spectrum | 15 (6%) | 11 (8%) | 4 (3%) | |
| PA-IVS | 9 (3%) | 6 (5%) | 3 (2%) | |
| DORV | 4 (1%) | 3 (2%) | 1 (1%) | |
| Other | 37 (15%) | 10 (8%) | 27 (22%) | |
| Number of previous surgical procedures, n (%) | 0.41 | |||
| 0 | 107 (43%) | 60 (47%) | 47 (37%) | |
| 1 | 71 (28%) | 34 (27%) | 37 (30%) | |
| 2 | 56 (22%) | 26 (20%) | 30 (24%) | |
| 3 | 18 (7%) | 7 (6%) | 11 (9%) | |
| Previous RV-PA conduit, n (%) | 71 (28%) | 54 (43%) | 17 (14%) | <0.001 |
| Current indication for conduit implantation, n (%) | <0.001 | |||
| Regurgitation | 81 (32%) | 22 (17%) | 59 (47%) | |
| RVOTO | 24 (10%) | 18 (14%) | 6 (5%) | |
| Mixed RVOTO and regurgitation | 22 (9%) | 16 (13%) | 6 (5%) | |
| Endocarditis | 3 (1%) | 2 (2%) | 1 (1%) | |
| Other (including primary surgical repair) | 122 (48%) | 69 (54%) | 53 (42%) | |
| Additional surgical procedures, n (%) | <0.001 | |||
| Rastelli/Truncus surgery | 18 (7%) | 18 (14%) | 0 (0%) | |
| Branch pulmonary plasty | 10 (4%) | 8 (6%) | 2 (2%) | |
| TV repair | 7 (3%) | 0 (0%) | 7 (6%) | |
| AV replacement | 3 (1%) | 1 (1%) | 2 (2%) | |
| AV repair | 2 (1%) | 1 (1%) | 1 (1%) | |
| MV replacement | 2 (1%) | 0 (0%) | 2 (2%) | |
| TV replacement | 2 (1%) | 0 (0%) | 2 (2%) | |
| Ross procedure | 2 (1%) | 2 (2%) | 0 (0%) | |
| MV repair | 1 (1%) | 1 (1%) | 0 (0%) | |
| Other | 34 (12%) | 26 (20%) | 8 (6%) | |
| Current implanted conduit type, n (%) | <0.001 | |||
| Pulmonary homograft | 98 (39%) | 39 (31%) | 59 (47%) | |
| Biopulmonic | 42 (17%) | 19 (15%) | 23 (18%) | |
| Contegra | 30 (12%) | 29 (23%) | 1 (1%) | |
| Matrix | 20 (8%) | 20 (16%) | 0 (0%) | |
| Carpentier | 16 (6%) | 5 (4%) | 11 (9%) | |
| Labcor | 6 (2%) | 6 (5%) | 0 (0%) | |
| Hancock | 2 (1%) | 0 (0%) | 2 (2%) | |
| Vascutek | 1 (1%) | 1 (1%) | 0 (0%) | |
| Other | 37 (15%) | 8 (6%) | 29 (23%) | |
| Conduit nominal diameter, mm | 22 (18-25) | 19 (16-22) | 24 (23-25) | <0.001 |
| Cardiopulmonary bypass time, minutes | 111 (83-148) | 124 (84-157) | 102 (80-140) | 0.02 |
| Cross-clamp time, minutes | 66 (50-96) | 80 (54-110) | 61 (41-78) | 0.001 |
| Fever within 7-days from surgery, n (%) | 52 (21%) | 36 (28%) | 16 (13%) | 0.002 |
| Large conduit (>22mm), n (%) | 155 (62%) | 42 (33%) | 113 (90%) | <0.001 |
| Long-term anticoagulation, n (%) | 33 (13%) | 8 (6%) | 25 (20%) | 0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
