Endovascular Stenting of Obstructed Right Ventricle–to–Pulmonary Artery Conduits
A 15-Year Experience
Background— The optimal treatment for dysfunctional right ventricle–to–pulmonary artery (RV-PA) conduits is unknown. Limited follow-up data on stenting of RV-PA conduits have been reported.
Methods and Results— Between 1990 and 2004, deployment of balloon-expandable bare stents was attempted in 242 obstructed RV-PA conduits in 221 patients (median age, 6.7 years). Acute hemodynamic changes after stenting included significantly decreased RV systolic pressure (89±18 to 65±20 mm Hg, P<0.001) and peak RV-PA gradient (59±19 to 27±14 mm Hg, P<0.001). There were no deaths, and, aside from 5 malpositioned stents requiring surgical removal, there were no serious procedural complications. During follow-up of 4.0±3.2 years, 9 patients died and 2 underwent heart transplantation, none related to catheterization or stent malfunction. During 155 follow-up catheterizations in 126 patients, the stent was redilated in 83 patients and additional stents were placed in 41. Stent fractures were diagnosed in 56 patients (43%) and associated with stent compression and substernal location but did not cause acute hemodynamic consequences. By Kaplan-Meier analysis, median freedom from conduit surgery after stenting was 2.7 years (3.9 years in patients >5 years), with younger age, homograft conduit, conduit diameter ≤10 mm, diagnosis other than tetralogy of Fallot, Genesis stent, higher prestent RV:aortic pressure ratio, and stent malposition associated with shorter freedom from surgery. Tricuspid regurgitation and RV function did not change between stent implantation and subsequent surgery.
Conclusions— Conduit stenting is an effective interim treatment for RV-PA conduit obstruction and prolongs conduit lifespan in most patients. Stent fractures were common but not associated with significant complications or earlier conduit reoperation.
- double-outlet right ventricle
- pulmonary valve
- tetralogy of Fallot
- transposition of the great arteries
- truncus arteriosis
Received October 14, 2005; de novo received December 8, 2005; revision received February 13, 2006; accepted March 13, 2006.
Standard treatment for many congenital heart defects includes placement of a conduit from the right ventricle (RV) to the pulmonary arteries (PA). The lifespan of RV-PA conduits is limited because of progressive obstruction and regurgitation. Freedom from replacement of homograft or bioprosthetic conduits ranges from 68% to 95% at 5 years and 0% to 59% at 10 years, with even shorter survival of small conduits.1–8 To prolong conduit lifespan and reduce the number of open heart surgeries patients undergo, we9–11 and others12–21 have attempted endovascular treatment before referral for surgery. These studies demonstrate that stenting obstructed RV-PA conduits with bare, nonvalved stents acutely reduces the peak pressure gradient across the conduit, RV systolic pressure, and the RV:aortic pressure ratio.11–14 Although smaller reports from our center and the Hospital for Sick Children in Toronto suggest that stent placement delays conduit reoperation for an average of 2.5 to 4 years,9–14 the clinical utility of bare stenting in these patients and factors associated with longer freedom from conduit reoperation require further definition. Furthermore, it has been observed that a proportion of RV-PA conduit stents fracture at sites of conduit obstruction, but contributing factors and clinical implications of stent fracture are unknown.
Editorial p 2569
Clinical Perspective p 2605
Improved understanding of these issues is particularly important now, with the advent of catheter-implanted valved stents. To assess the expected duration of benefit from conduit stenting, factors associated with greater benefit, and predictors and clinical implications of stent fracture, we reviewed our 15-year experience with RV-PA conduit stenting.
Data from patients who underwent placement of one or more bare stents (ie, nonvalved, noncovered stents) in a pulmonary outflow conduit at Children’s Hospital between January 1990 and December 2004 were reviewed. Patients with corrected transposition of the great arteries who underwent stenting of a conduit from the morphological left ventricle to the PAs were included. Because some patients had multiple different conduits stented or multiple stents placed in a single conduit, data were tabulated and analyzed with stented conduits as the index unit (except for mortality, which was analyzed with patients as the index unit). Additional stents placed in the same conduit at a subsequent catheterization were analyzed separately and are discussed in a section on follow-up catheterization.
The technical approach to conduit stenting has been described previously.11 After hemodynamic and angiographic assessment of conduit obstruction and valve competence, aortic root or selective coronary angiography was performed to assess the proximity of the coronary arteries to the conduit. In most cases, the conduit stenosis was predilated with one or more balloons up to 110% of the nominal conduit diameter (ie, balloon:nominal conduit diameter ratio 1.1). In patients with coronary arteries passing close to the conduit, aortic root or coronary angiography was performed with a balloon inflated across the conduit stenosis to evaluate the potential for coronary compression by a conduit stent. If coronary compression was noted, stenting was not performed. Otherwise, a stent was mounted on a balloon of the appropriate size and advanced across the area of stenosis either through a previously placed long sheath or front-loaded into a long sheath and advanced over a wire (alternatively, a premounted stent was chosen). The stent was expanded and further dilated as indicated. Balloon-expandable Palmaz stainless steel stents (Cordis Endovascular, Warren, NJ) were used until June 2002 when Palmaz Genesis stents (Cordis Endovascular, Miami, Fla) became available, including premounted versions. Stent length was selected on an individual basis after predilation of the obstruction. In patients with a functioning conduit valve, based on the presence of a diastolic gradient between the PAs and RV, every effort was made to deploy the stent without compromising the mobility of the conduit valve leaflets.
Hemodynamic data were obtained from echocardiograms and catheterizations. Prestent and poststent MRI data were available for a small minority of patients and are not presented in this report. The severity of RV dysfunction and the degree of tricuspid (TR) and pulmonary (PR) regurgitation were assessed by echocardiography (or angiography if adequate echocardiographic images were not available) and categorized qualitatively as either none/mild or moderate/severe. If different Doppler gradients across the RV-PA conduit were measured from different windows, the highest reported gradient was used for analysis. In patients with conduit and branch PA obstruction, the calculated RV-PA pressure gradient at catheterization was based on the PA pressure distal to the conduit obstruction but proximal to the PA stenosis.
Angiograms from the index catheterization and all follow-up angiograms were reviewed. Conduit calcification was graded as severe (heavy, visible along entire conduit), mild (minimally visible along a portion of the conduit), or none. The diameter of the stenotic segment of conduit was measured in the anteroposterior and lateral projections before and after stent placement. To determine the degree of asymmetry of the expanded stent, typically in the setting of compression between the heart and anterior chest wall, the ratio of lateral-dimension diameter to anteroposterior-dimension diameter was calculated. The absolute value of the difference between this ratio and unity was calculated as a “compression index,” which reflects compression/asymmetry in either the anteroposterior or lateral dimension. The location of the stent was characterized as substernal (stent behind and directly apposed to anterior chest wall), partially substernal (&50% or less of stent apposed to anterior chest wall), or remote from the chest wall.
Patients who underwent follow-up catheterization, chest fluoroscopy, or conduit surgery were assessed for gross stent fractures. All follow-up angiograms were reviewed to assess for stent fracture. Because stent fractures may only be evident with dynamic imaging or direct inspection, simple chest radiography was not considered sufficient to exclude the possibility of stent fracture. Accordingly, evaluation of this variable was performed as an analysis of freedom from diagnosis of stent fracture. Stent fractures were characterized according to orientation (longitudinal, transverse, compound), severity (fragmented, complete length or circumference, partial length or circumference) and preserved structural integrity.
The primary outcome assessed was freedom from surgical conduit reintervention after conduit stenting. Secondary outcomes included acute changes in hemodynamic variables, worsening of RV dysfunction or TR, diagnosis of stent fracture, and transplant-free survival. Independent variables analyzed for association with outcomes included age, diagnosis, year of conduit stenting, duration between conduit placement and stenting, conduit type, conduit size, prior conduit augmentation or dilation, degree of conduit calcification, stent type, diameter of largest dilating balloon, prestent and poststent hemodynamic variables, changes in right-heart hemodynamic variables after stent placement, stent compression, substernal stent location, interventions on the branch PAs at the time of conduit stenting, and acute complications. Time-dependent outcomes (freedom from conduit surgery, transplant-free survival, freedom from diagnosis of stent fracture) were assessed with the Kaplan-Meier product-limit method. Factors associated with time-dependent outcomes were analyzed with the use of Cox proportional hazards regression. Factors significant by univariable Cox regression analysis were entered into a multivariable model by using Forward stepwise entry. Analysis of freedom from diagnosis of stent fracture included only patients who had undergone follow-up catheterization or fluoroscopy, and patients were censored event-free at the time of conduit surgery. Although stent fracture should be considered a time-dependent outcome, it was generally not clinically evident and depended on assessment with catheterization or surgery, making time-variant analytic methods possibly suboptimal. Accordingly, we analyzed factors associated with diagnosis of stent fracture by using both Cox regression and logistic regression analysis. For comparison of preintervention and postintervention hemodynamic indexes, paired t-test analysis was used. For analysis of factors associated with acute outcome and between-group comparisons of means and proportions, either independent-samples t test or 1-way ANOVA and either χ2 analysis or the Fisher exact test were used, respectively. Correlation between continuous variables was assessed by using linear regression analysis. Data are presented as mean±SD or median (range).
The study was approved by the Children’s Hospital Institutional Review Board. The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
From January 1990 through December 2004, cardiac catheterization was performed and stent deployment was attempted in 242 obstructed pulmonary outflow conduits in 221 patients (2 separate conduits were stented in 21 patients), including 44 previously reported patients.11 Demographic and conduit-related data are summarized in Table 1. The median age at the time of original conduit placement was 1.7 years (3 days to 41 years). The median age at the time of conduit stenting was 6.7 years (0.3 to 48 years), and a median of 3.7 years (0.1 to 26 years) had elapsed between conduit placement and stenting. Twenty-one patients (9%) had prior surgical augmentation of the conduit and 16 (7%) had undergone prior balloon dilation of the conduit without stenting. A native ventricular septal defect was open (n=13) or partially closed with a fenestrated patch (n=4) in 17 patients (7%, not including unintentional residual or small muscular ventricular septal defects).
Catheterization and Conduit Stenting
At the time of the index catheterization, 270 stents were successfully placed in 241 conduits, with multiple overlapping or serial stents placed in 28 conduits (2 stents in 27 conduits, 3 stents in 1). One attempted stent deployment was unsuccessful because of entrapment of the front-loaded stent in the tricuspid valve (TV) during advancement through the RV. Deployed stent types included Palmaz stents (n=214), premounted Palmaz Genesis stents (n=25), and nonpremounted Palmaz Genesis stents (n=31). From June 2002, Palmaz Genesis stents comprised all but 5 of the conduit stents placed. Among the 270 implanted stents, length was 10 to 15 mm in 34, 18 to 20 mm in 110, 25 to 30 mm in 106, and >30 mm in 20.
The conduit was predilated in 220 of 242 cases (91%). In 1 additional case, predilation of the conduit with simultaneous coronary angiography demonstrated obstruction of the coronary artery by the balloon, and in another, coronary angiography demonstrated close proximity between the left main coronary artery and the underside of the conduit, so no stent was implanted in either case. The maximum balloon size used for dilation of the stent ranged from 5.5 mm to 26 mm and was <10 mm in 29 patients, 10 to 12 mm in 71, 14 to 16 mm in 95, 17 to 19 mm in 31, and ≥20 mm in 15. Among patients who had not undergone prior conduit augmentation, the ratio of starting balloon diameter to the nominal conduit diameter was 0.82±0.21, and the maximum balloon:nominal conduit diameter ratio was 0.97±0.21. The maximum balloon:nominal conduit diameter ratio exceeded 1.1 in 14% of conduits (28 of 203) and was significantly more likely in conduits with a nominal diameter ≤10 mm (35% versus 8%, P<0.001).
Dilation or stenting of the branch PAs was performed during the same catheterization in 90 patients (37%). Patients with tetralogy of Fallot (TOF) were significantly more likely to undergo concurrent PA interventions than patients with other diagnoses (47% versus 30%, P=0.01).
On prestent imaging, 29 of 221 patients (13%) with adequate images for evaluation had moderate/severe RV dysfunction, 18 of 221 patients (8%) had moderate/severe TR, and 162 of 210 patients (77%) had moderate/severe PR. The estimated maximum instantaneous gradient across the RV-PA conduit by Doppler analysis was 64.7±19.4 mm Hg. This was significantly higher than (P<0.001) and correlated poorly with (R2=0.06) the gradient obtained at catheterization (Table 2), with systematically higher measurements by Doppler and no apparent trend by Bland-Altman analysis.
Hemodynamic data measured at catheterization, before and after stent placement, are summarized in Table 2. The only patient-related or procedural variables associated with a larger decrease in RV pressure after stent placement were a higher prestent RV pressure (r=0.47, P<0.001), diagnosis other than TOF (P=0.03), and intervention on the branch PAs at the same procedure (P<0.001). The only variables that correlated with a greater decrease in RV-PA pressure gradient were a higher prestent RV-PA pressure gradient (r=0.71, P<0.001) and intervention on the branch PAs at the same procedure (P<0.001).
Acute and Subacute Adverse Events
There were no deaths, strokes, myocardial infarctions, cardiac perforations, or damage to the conduit requiring surgical treatment. Including the patient in whom the stent became entrapped in the TV during attempted advancement to the conduit, 5 procedures (2.1%) were complicated by acute or subacute stent malposition/embolism to the TV or RV requiring surgical removal. Since January 2000, only 1 of 100 cases was complicated by stent malposition requiring surgical removal (1%), compared with 4 of the previous 142 cases (3%, P=0.3). In another 5 patients (2.1%), there was stent malposition, with successful endovascular management and ultimate deployment in the PAs or central systemic venous system, with an additional stent placed for treatment of the conduit obstruction. Patients in whom predilation of the conduit was not performed were significantly more likely to experience stent malposition than those in whom predilation was performed (3 of 22 patients, compared with 7 of 220 patients, P=0.05).
In 6 patients, there was extravasation of contrast through a conduit crack/tear after conduit dilation and/or stenting, none of whom developed hemothorax or hemopericardium that required drainage or surgery. In 1 of these patients, a covered stent was placed across the tear, and in the other 5 the extravasation was self-limited and no longer evident on repeat angiography. Fluoroscopically visible fractures of the calcified conduit layer without extravasation of contrast were common. One or more balloons ruptured during dilation and/or stent deployment in 74 cases (30%). In all but 3 of these cases, which required venous cutdown, the balloon fragments were removed through the venous sheath or an additional venous sheath (n=6) placed for the purpose of snaring and removing the balloon.
Follow-up data were available for all but 8 patients (97%), with a mean follow-up duration of 4.0±3.2 years.
Death or Heart Transplantation
There were no acute deaths after conduit stenting. During follow-up, 11 patients died (n=9) or underwent a heart transplantation (n=2) a median of 1.6 years after conduit stenting. In 7 of these patients, the stented conduit had been replaced. Causes of death included postoperative complications after conduit replacement (n=3), severe pulmonary hypertension (n=1), cardiomyopathy leading to progressive biventricular dysfunction (n =1), severe aortic regurgitation leading to hypotension and cardiac arrest (n=1), sudden death (n=1), accident (n=1), and sepsis (n=1). Indications for heart transplantation included cardiac arrest as the result of coronary artery thromboocclusion after surgical conduit replacement, without return of ventricular function during prolonged extracorporeal circulatory support (n=1), and progressive biventricular dysfunction (n=1). A pediatric cardiac surgeon and noninterventional pediatric cardiologist reviewed the records of these 11 patients and determined that probably none of the deaths or transplantations were attributable to the conduit stent or stenting procedure. By Kaplan-Meier analysis, transplant-free survival after conduit stenting was 99% at 1 year, 95% at 5 years, and 90% at 10 years.
A total of 155 follow-up catheterizations (before any conduit surgery) were performed in 126 patients. The first follow-up catheterization was performed 2.2±1.7 years after index stent placement. The RV pressure (86.4±21.1 mm Hg, P=0.4), peak RV-PA conduit gradient (53.8±21.6 mm Hg, P=0.18), and RV:aortic pressure ratio (0.85±0.19, P=0.09) had all returned nearly to prestent levels.
Current or preoperative (in patients who underwent conduit reoperation) follow-up echocardiograms with adequate images available for review of RV function were available in 181 patients, and of TR in 196 patients, 2.7±2.3 years after conduit stenting. Among the 29 patients with moderate RV dysfunction before stenting, 17 had moderate RV dysfunction before conduit replacement, 5 had mild RV dysfunction, and 7 did not have adequate images on follow-up/preoperative imaging (1 of these 7 died 1 month after stenting and 1 underwent conduit reoperation 2 months after stenting). Among 192 patients with no/mild RV dysfunction before stent placement, 4 (2%) had moderate/severe RV dysfunction before conduit replacement, and 30 others did not have adequate follow-up images to evaluate RV function (16 followed up for <1 year, including 8 who underwent conduit replacement within 1 year of stenting). Four patients who did not have adequate prestent echocardiograms available for review of RV dysfunction had moderate/severe RV dysfunction on follow-up echocardiography. No factors were identified that were associated with progression of RV dysfunction.
Of the 18 patients with moderate or severe TR before conduit stenting, 8 had improved to mild TR at follow-up, 9 remained moderate or severe, and 1 did not have adequate images available for determination of TR severity. Of the 203 patients with no/mild TR before conduit stenting, 2 progressed to moderate/severe (1%), 164 remained unchanged, and 37 did not have adequate echocardiographic images. Of the 2 patients with progressive TR, 1 also had worsening RV function.
Of the 48 patients with less than moderate PR before conduit stenting, 29 progressed to moderate/severe PR (2.5±1.8 years between stenting and follow-up echocardiogram), 8 had unchanged PR, and 11 did not have adequate echocardiographic images of the conduit.
During follow-up catheterization (155 procedures in 126 patients), 87 patients underwent 100 repeat balloon dilations of the conduit stent, and 41 of these 81 patients had 1 (n=33) or more (n=8) additional conduit stents placed (total, 50 stents). Among patients undergoing repeat dilation of the conduit stent or placement of an additional stent, there were significant reductions compared with baseline in RV systolic pressure, RV-PA pressure gradient, and RV:aortic pressure ratio (Table 2).
Follow-Up Conduit Surgery
A total of 146 patients underwent surgical reintervention on the RV-PA conduit. The conduit was replaced in 108 patients and augmented with a patch in 38 patients. By Kaplan-Meier analysis, the median freedom from conduit surgery after stent placement was 2.7 years overall and 3.9 years in patients older than 5 years of age at the time of stenting. Eighty-three of these 146 patients had undergone poststent catheterization, a median of 5 months before conduit surgery, at which time RV pressure, the peak RV-PA conduit gradient, and RV:aortic pressure ratio were not significantly different than immediately before stent placement (all P>0.3). There were 3 perioperative deaths (2%) after conduit replacement/revision (see above).
Factors associated with longer freedom from conduit surgery after conduit stenting are summarized in Table 3. By multivariable Cox regression, factors independently associated with shorter freedom from conduit surgery after stenting included younger age, higher prestent RV:aortic systolic pressure ratio, diagnosis other than TOF, homograft conduit, nominal conduit diameter ≤10 mm, use of a Genesis stent, and malposition of the conduit stent (Figure 1 and Table 2).
Stent fractures were identified in 56 patients, including 54 of 126 patients (43%) who underwent repeat catheterization during follow-up and 2 of 63 patients who underwent conduit surgery without preceding follow-up catheterization (overall 30% of these 189 patients). Fifty of 56 stents that fractured (89%) were located immediately behind the sternum or chest wall and were compressed between the chest wall and heart. Fractures were longitudinal or compound in 51 cases (Figure 2) and transverse in 5. The integrity of the fractured stent was compromised in most cases (46 of 56, 82%), although partial fractures (1 to 3 cells) were observed in 10 stents and did not result in loss of stent integrity. Stent fragments embolized to the RV and/or PA in 14 cases (Figure 2). In 2 of the cases with an embolized fragment, there was a complete transverse stent fracture, with embolization of the proximal segment to the RV (n=1) or of the distal segment to the right PA (n=1). These patients did not have any associated symptoms. The stent fracture was not reported (not diagnosed) in the original catheterization report in 10 cases. In no cases did the stent fracture cause or contribute to death or acute hemodynamic consequences.
By multivariable Cox regression, a higher compression index (B=4.3, P<0.001), substernal stent location (B=−2.3, P=0.002), and use of a Genesis stent (B=−1.4, P=0.006) were independently associated with shorter freedom from diagnosis of stent fracture. When the analysis was repeated with diagnosis of stent fracture as a non–time-dependent outcome, using logistic regression analysis, findings were the same.
Twenty-three patients had additional stents placed within the fractured stent. Forty-two patients underwent conduit reoperation a median of 3 months after diagnosis of the stent fracture, but stent fracture was not associated with shorter freedom from conduit surgery after stent placement by Kaplan-Meier analysis.
Efficacy of Stent Placement for RV-PA Conduit Obstruction
In our experience and that of others,12–14 treatment of RV-PA conduit obstruction with endovascular stenting decreases RV pressure and RV-PA obstruction acutely and prolongs conduit lifespan in most patients without an increased risk of deterioration in RV function. Over the past 15 years, we have successfully stented 241 obstructed RV-PA conduits, with good acute hemodynamic results and few significant complications. Acute reductions in RV pressure and RV-PA pressure gradient were more pronounced in patients with higher prestent pressure/gradient and in those who did not undergo concurrent interventions on the branch PAs. By Kaplan-Meier analysis, the median freedom from conduit surgery after conduit stenting was 3.9 years in children older than 5 years and 2.7 years overall. By the time of conduit operation, RV pressure and conduit gradients had returned to prestent levels, presumably as the result of a combination of factors, including patient growth, neointimal ingrowth, and conduit/stent compression. The next largest reported experience of RV-PA conduit stenting, which includes 68 patients, describes a median freedom from conduit replacement after stenting of more than 4 years, which is longer than in our experience.14 However, it is difficult to compare results between series, given the high degree of variability in the timing of conduit intervention. For example, in the series by Sugiyama et al14 the median RV pressure and conduit gradient at the time of conduit stenting were 66 mm Hg and 44 mm Hg, respectively, compared with 89 mm Hg and 58 mm Hg in our cohort.
The most important potential downsides of RV-PA conduit stenting are procedural mortality/morbidity and deterioration of RV function caused by persistent pressure and volume overload during the interval between stenting and eventual conduit surgery. In our experience, there were no deaths associated with catheterization or stenting, and serious complications were rare, consisting exclusively of stent embolization/malposition requiring surgery. There was no demonstrable risk of deterioration in RV and/or TV function after conduit stenting: Only 5 of 165 patients with adequate images had development of moderate TR or RV dysfunction during follow-up, whereas 13 patients with at least moderate TR or RV dysfunction before stenting improved.
A novel finding of this study was the high frequency of stent fracture after RV-PA conduit stenting. Over 40% of patients who underwent follow-up catheterization were diagnosed with a fractured stent, which may even underestimate the true incidence, as small fractures can be difficult to visualize and intraoperative evaluation may not have detected stent fractures if the conduit was excised completely without first being opened. Stent integrity was compromised in most cases, and stent fragments embolized to the RV or PA in 14 of 56 fractured stents. Although unexpected, it is not surprising that RV-PA conduit stents fracture so often. In many cases, and most of those in which fractures were observed in our series, the obstructed portion of the conduit is immediately behind the anterior chest wall, and the stent is effectively compressed between the chest wall and the heart. Bench-testing of Palmaz and Genesis stents demonstrates relatively high radial strength,22–24 but vascular stents are not necessarily suited to withstand the asymmetric, cyclic compressive stresses to which they are subject in the setting of conduit compression between the chest wall and the beating heart. Stents used in other locations exposed to compression or cyclic stress are also observed to fracture relatively often.25,26
With respect to stent fracture, another concerning finding of this study is that Palmaz Genesis stents, which have replaced Palmaz stents in the product line and in our practice, had a significantly greater tendency than Palmaz stents to fracture when deployed in obstructed RV-PA conduits. Radial strength testing with application of full-length area-loading has shown Palmaz and Palmaz Genesis stents to have similar radial strength,22 but the discrepancy between these in vitro data and our clinical experience suggests that radial strength testing by full-length area-load methods does not accurately simulate the stresses to which a stent is exposed in an RV-PA conduit.
Although the high rate of stent fracture in RV-PA conduits is concerning, it is reassuring that stent fracture did not lead to hemodynamic compromise or other significant adverse events and was not associated with shorter freedom from conduit surgery. The implications of our findings with regard to stent fracture, which require corroboration and further study, are not yet clear. Routine chest fluoroscopy during follow-up may lead to earlier diagnosis of stent fractures, particularly in patients with a substernal stent that appears to be at higher risk for compression.
Predilation of the stenotic area is important for safe and effective stenting of RV-PA conduits. Compliance of the conduit is unpredictable, and predilation allows for determination of the exact location of the waist, the presence of multiple waists, and noncompliant segments that may displace the balloon during inflation and result in stent malposition. Of note, there was a higher probability of stent malposition among the small subset of patients in our series who did not undergo predilation before stenting. Another benefit of predilation is that it allows for determination of potential coronary artery compression if performed with simultaneous aortic root or coronary angiography.
Conduit dysfunction involves components of both obstruction and regurgitation, and different individuals will not tolerate similar degrees of conduit dysfunction in the same way. It is difficult to assess the effectiveness of treatment for conduit dysfunction and the clinical implications of conduit treatment in a standardized manner, particularly in a retrospective evaluation. More sensitive functional analyses available with MRI or clinical evaluation such as exercise testing, which may have helped evaluate the progress of patients after conduit stenting, were not routinely available in our cohort. Also, coexisting branch PA stenoses and interventions, which were common in our patients, may affect the efficacy of conduit stenting. Patients who underwent concurrent branch PA interventions had less acute reduction in RV pressure and RV-PA pressure gradient, but no difference in freedom from conduit surgery, compared with patients who did not undergo branch PA dilation or stenting.
Valved RV-PA Conduit Stents
A drawback of bare stenting for treatment of RV-PA conduit dysfunction is that it does not improve conduit regurgitation, which is present in the majority of patients with conduit obstruction (77% of our patients had at least moderate PR). Valved stents are a novel therapy for both RV-PA conduit obstruction and regurgitation that has been reported in 59 patients, with encouraging early results.27 The data presented in this report will serve as the largest comparison series of bare stenting for RV-PA conduit obstruction, with the longest follow-up, and may help anticipate important issues with this emerging technology, such as stent fracture and early restenosis.
Sources of Funding
This study was supported in part by a generous contribution from Todd and Marta Loeb.
Mohammadi S, Belli E, Martinovic I, Houyel L, Capderou A, Petit J, Planche C, Serraf A. Surgery for right ventricle to pulmonary artery conduit obstruction: risk factors for further reoperation. Eur J Cardiothorac Surg. 2005; 28: 217–222.
Dearani JA, Danielson GK, Puga FJ, Schaff HV, Warnes CW, Driscoll DJ, Schleck CD, Ilstrup DM. Late follow-up of 1095 patients undergoing operation for complex congenital heart disease utilizing pulmonary ventricle to pulmonary artery conduits. Ann Thorac Surg. 2003; 75: 399–410.
Levine AJ, Miller PA, Stumper OS, Wright JG, Silove ED, De Giovanni JV, Sethia B, Brawn WJ. Early results of right ventricular-pulmonary artery conduits in patients under 1 year of age. Eur J Cardiothorac Surg. 2001; 19: 122–126.
Karamlou T, Ungerleider RM, Alsoufi B, Burch G, Silberbach M, Reller M, Shen I. Oversizing pulmonary homograft conduits does not significantly decrease allograft failure in children. Eur J Cardiothorac Surg. 2005; 27: 548–553.
Boethig D, Thies WR, Hecker H, Breymann T. Mid term course after pediatric right ventricular outflow tract reconstruction: a comparison of homografts, porcine xenografts and Contegras. Eur J Cardiothorac Surg. 2005; 27: 58–66.
O’Laughlin MP, Perry SB, Lock JE, Mullins CE. Use of endovascular stents in congenital heart disease. Circulation. 1991; 83: 1923–1939.
O’Laughlin MP, Slack MC, Grifka RG, Perry SB, Lock JE, Mullins CE. Implantation and intermediate-term follow-up of stents in congenital heart disease. Circulation. 1993; 88: 605–614.
Powell AJ, Lock JE, Keane JF, Perry SB. Prolongation of RV-PA conduit life span by percutaneous stent implantation. Circulation. 1995; 92: 3282–3288.
Sugiyama H, Williams W, Benson LN. Implantation of endovascular stents for the obstructive right ventricular outflow tract. Heart. 2005; 91: 1058–1063.
Hatai Y, Nykanen DG, Williams WG, Freedom RM, Benson LN. Endovascular stents in children under 1 year of age: acute impact and late results. Br Heart J. 1995; 74: 689–695.
Fogelman R, Nykanen D, Smallhorn J, McCrindle BW, Freedom RM, Benson LN. Endovascular stents in the pulmonary circulation: clinical impact on management and medium-term follow-up. Circulation. 1995; 92: 881–885.
Forbes TJ, Rodriguez-Cruz E, Amin Z, Benson LN, Fagan TE, Hellenbrand WE, Latson LA, Moore P, Mullins CE, Vincent JA. The Genesis stent: a new low-profile stent for use in infants, children, and adults with congenital heart disease. Catheter Cardiovasc Interv. 2003; 59: 406–414.
Kalmar G, Jubner F, Wolfram V, Hutzenlaub J, Teubner J, Poerner T, Suselbeck T, Borggrefe M, Haase KK. Radial force and wall apposition of balloon-expandable vascular stents in eccentric stenoses: an in vitro evaluation in a curved vessel model. J Vasc Interv Radiol. 2002; 13: 499–508.
Khambadkone S, Coats L, Taylor A, Boudjemline Y, Derrick G, Tsang V, Cooper J, Muthurangu V, Hegde SR, Razavi RS, Pellerin D, Deanfield J, Bonhoeffer P. Percutaneous pulmonary valve implantation in humans: results in 59 consecutive patients. Circulation. 2005; 112: 1189–1197.
Standard treatment for many congenital heart defects includes placement of a conduit from the right ventricle to the pulmonary arteries. The lifespan of these conduits is limited by progressive obstruction and regurgitation. Studies have demonstrated that endovascular treatment involving stenting of obstructed conduits delays surgical reintervention, but little is known about factors contributing to longer freedom from reoperation. Our study demonstrates that freedom from conduit surgery after stent placement was 2.7 years overall and 3.9 years in patients older than 5 years of age at the time of stenting. Stenting acutely reduces RV pressure, is not related to any procedural deaths or serious morbidity, and preserves RV and TV function. Although stent fractures occur frequently, they are asymptomatic and do not affect freedom from conduit reoperation. Factors independently associated with longer freedom from surgical reintervention include older age, lower prestent RV:aortic systolic pressure, nominal conduit diameter >10 mm, non-Genesis stents, diagnosis of TOF, and nonhomograft conduits. These findings can help the practicing cardiologist gauge when a patient may need to have surgical reintervention after stenting of an obstructed RV-PA conduit. Furthermore, the present study may be used for comparative assessment of outcomes in patients undergoing placement of valved stents in the RV outflow tract.
Guest Editor for this article was Robyn J. Barst, MD.