“Hybrid” Stent Delivery in the Pulmonary Circulation

ABSTRACT: Objective. To describe our institutional experience providing “hybrid” intraoperative stent therapy for pulmonary artery (PA) stenoses. Background. Surgical patch angioplasty, transcatheter stent therapy and intraoperative stent delivery are valuable treatment options for PA stenoses. The experience with intraoperative hybrid therapy has increased and new techniques and equipment have become available. Methods. This study involves a retrospective review of 20 patients with a median age of 5.1 years who underwent hybrid PA stent therapy between March 2003 and April 2008. Thirteen patients had underlying diagnoses of either tetralogy of Fallot, pulmonary atresia with a ventricular sepal defect (VSD) or truncus arteriosus. Results. In 15 procedures, stents were implanted under direct vision. In 3 procedures, stents were implanted using vascular puncture with angiographic guidance, and in 2 procedures, stents were implanted using a combination of direct vision and fluoroscopy/angiography. Successful stent deployment was achieved in 18/20 (90%) procedures. One stent was malpositioned from the right ventricular outflow tract through a VSD, while another stent fractured as a result of high-pressure balloon expansion of a very resistant lesion. Adverse events were encountered in 3/20 (15%) procedures, which included the 2 “unsuccessful” stent deliveries. The median follow up thus far has been 1.7 years (41 days – 5.1 years). Seven of 20 (35%) patients required repeated interventions at the site of previous stent placement at a median interval of 8.5 months. Conclusions. Hybrid stent delivery in the operating room, using either direct vision or vascular puncture, is safe and effective. Using a well-equipped hybrid operating suite facilitates safe intraoperative stent delivery in a wide variety of patients. Close cooperation between the surgical and interventional teams is essential. J INVASIVE CARDIOL 2008;20:592–598 Treatment of pulmonary artery (PA) stenoses continues to remain a challenge for patients with congenital heart disease, particularly those with tetralogy of Fallot, truncus arteriosus, as well as pulmonary atresia and ventricular septal defects (VSD). Surgical patch angioplasty has long been considered the gold standard, but results have frequently been disappointing, which has led to the evaluation of other treatment strategies. Management strategies have evolved over time, and the introduction of endovascular stents to the interventional armamentarium of the congenital cardiologist, pioneered by Chuck Mullins,1–3 has proven the benefits of endovascular stent therapy. However, stent delivery of adult-sized stents frequently requires large delivery systems, is technically demanding and not without complications. Delivery of endovascular stents in the operating room (OR) combines the best of both worlds. An increase in operator experience, combined with an increasingly cooperative environment between cardiac surgeons and congenital interventionalists, has led to expansion and improved results of this unique hybrid approach. We describe our institutional experience using hybrid stent delivery for pulmonary artery rehabilitation. Methods Study population. The study was conducted as a retrospective case note review and institutional review board approval was obtained. Local procedures for retrospective record review and patient confidentiality were followed. All hybrid procedures that included stent placement in a pulmonary artery through procedural cooperation between cardiothoracic surgeon and cardiac interventionalist were identified using an electronic procedural database maintained in Microsoft Access. Between March 2003 and April 2008, 20 patients underwent hybrid stent deployment in a pulmonary artery, with placement of 27 endovascular stents to treat 24 vascular pulmonary artery lesions. Seventeen of 20 (85%) procedures were performed in a conventional cardiac surgical OR, while 2 procedures were performed in a dedicated hybrid cardiac operating suite, and 1 procedure in a dedicated hybrid catheterization suite. The median age was 5.1 years (8 days–42 years). Sixteen of 20 (80%) patients had undergone cardiac catheterization up to 6 months before the hybrid intervention. Nine procedures were planned as elective periprocedural stent implantations (Figures 1, 2 and 3), 4 procedures were considered to possibly require periprocedural stent implantation (Figure 4), while 7 procedures were unplanned hybrid interventions resulting from unexpected intraoperative surgical findings (Figures 5 and 6). Four procedures resulted from previous failed attempts at transcatheter pulmonary artery rehabilitation (Figure 2). Underlying diagnoses included repaired tetralogy of Fallot (n = 7), repaired pulmonary atresia with VSD (n = 3), repaired truncus arteriosus (n = 3), complex 2-ventricle anatomy (n = 3), single ventricle post Glenn (n = 1), and other complex congenital heart diseases (n = 3). Surgical procedures performed in the same setting included conduit replacement or pulmonary valve replacement (PVR) (n = 11), biventricular repair of congenital heart disease (CHD) (n = 1), Fontan completion (n = 1), and unifocalization of major aortopulmonary collateral arteries (MAPCAs) (n = 1). In 5 procedures, surgical access was provided solely for completion of the hybrid intervention (Figures 1 and 2). Additional hybrid procedures performed in the same setting included balloon angioplasty of an in situ right PA stent (n = 1), cutting balloon angioplasty of a branch PA stenosis (n = 3), as well as balloon angioplasty of a right middle-lobe branch PA (n = 1). Data collection and outcome parameters. Data were collected retrospectively including demographic variables, underlying diagnosis, as well as preceding and subsequent surgical and transcatheter procedures. Imaging data (computed tomography [CT], magnetic resonance imaging [MRI], angiography, chest X-ray) obtained following stent placement and during the follow-up period were reviewed to assess for stent malposition, stent fracture, in-stent stenosis, jailing of pulmonary arterial branches or inappropriate stent size. Surgical as well as interventional details (technique, equipment used, type of approach) were documented for each procedure. Primary outcome parameters were procedural success, defined as placement of an endovascular stent in the desired position and expanded to the intended diameter, without evidence of malposition, stent migration or stent fracture, the need for reintervention or subsequent surgical stent removal, as well as the incidence of procedure-related adverse events. Statistical analysis. Basic descriptive variables (mean, median, standard deviation, range) as well as all statistical tests were calculated for continuous variables using StatsDirect software (StatsDirect Ltd., Cheshire, United Kingdom). Results In 12/20 (60%) procedures, stents were implanted in the LPA, in 1 procedure, stents were implanted in the RPA, and in 4 procedures, stents were implanted in both the LPA and RPA. Lobar pulmonary arterial branches were stented in 1 procedure and stents were placed across the right ventricular outflow tract (RVOT) in 2 procedures. In 15 procedures, stents were implanted under direct vision without the use of fluoroscopy or angiography, using endoscopic guidance in 9 procedures (Figure 4). A combination of direct vision and fluoroscopy/angiography was used in 2 procedures (Figure 5), while fluoroscopy/angiography alone was used in 3 procedures (Figure 1). After stent deployment, the result was assessed using exit angiography in 7 procedures, all of which documented appropriate stent size and position. Fluoroscopic guidance, with or without angiography, was obtained using either a conventional C-arm (n = 3 procedures), or a specifically-designed hybrid operating (Figure 6) or catheterization suite (n = 2 procedures). External compression was identified to be present in 2 vascular lesions, and a kink of a proximal branch PA in 6 lesions, usually the LPA (Figure 4). Implanted stents included the Mega LD (ev3, Inc., Plymouth, Minnesota) (n = 1), the Max LD (ev3) (n = 6), the Genesis XD (Cordis Corp., Miami Lakes, Florida) (n = 9), the covered ICast (Atrium, Hudson, New Hampshire) (n = 1), the ITI Double Strut (n = 1), and the premounted Genesis stent (Cordis) (n = 8). Successful stent deployment was achieved in 18/20 (90%) procedures. In 1 patient with PA VSD, a blind ending atretic pulmonary valve was perforated intraoperatively and a stent was placed through the RVOT. Angiography 1 week later showed that this stent did not extend into the RVOT, but instead protruded through the VSD into the left ventricle (Figure 7). The patient subsequently underwent placement of a right modified BT-shunt (RMBTS), and the stent was surgically removed during full correction at 9 months of age. In another patient, the stented lesion required pressures in excess of 20 atm to fully expand the stent, which led to stent fracture. However, the pressure gradient was successfully relieved. The fractured stent was subsequently removed 8 months later during elective surgical conduit replacement. Jailing of a side branch was not observed in any of the procedures. Adverse events. Overall, adverse events were encountered in 3/30 (15%) procedures, including the aforementioned patients with stent malposition and stent fracture, all of which occurred within a conventional cardiac OR. In addition, 1 premature 1.6 kg patient with pulmonary atresia with intact ventricular septum (PAITS) had a decrease in saturation during initial manipulation of the patent ductus arteriosus and the RVOT. This was exaggerated during stent deployment and was eventually improved after a combination of volume and calcium was administered to increase systemic pressure. Follow Up and Reintervention. The median follow-up period thus far has been 1.7 years (41 days–5.1 years). Seven of 20 (35%) patients required reinterventions at the site of the previous stent placement at a median interval of 8.5 months (6.9 months–1.2 years). These included transcatheter reinterventions performed in 4 patients, 2 of whom underwent stent expansion for interval growth, and 2 who required additional stent placement and/or balloon angioplasty to treat in-stent stenosis. Surgical procedures were performed in 3 patients, 2 of whom underwent elective removal of a RVOT stent during subsequent surgical repair, and 1 who underwent conduit replacement as well as bilateral stent removal with surgical PA angioplasty. An additional 3 patients underwent stent implantations in pulmonary artery positions that may exceed the maximum expandable stent diameter with growth. However, in 2 of these patients, stents were implanted in an extremely hypoplastic branch PA that may never exceed the maximum expandable diameter of these stents (Figures 2 and 3). Overall, only 1 case of stent fracture and 2 cases of in-stent stenosis were observed during the follow-up period (see above), but these data are limited by only 10/20 (50%) patients having undergone CT, MRI or angiographic evaluation during the follow-up period. Discussion Hybrid therapy has come a long way from a “surgical bailout procedure” to being considered an important alternative treatment strategy in the management of patients with congenital heart disease.4,5 What applies to hybrid therapy in general, equally applies to intraoperative stenting of the PAs, and the techniques available to perform intraoperative stent placement have expanded considerably.6–9 It is therefore a broad technical misconception that intraoperative stent placement involves crimping a stent on a balloon catheter and inflating it within a branch PA, an approach that easily leads to unsatisfactory results that are wrongly attributed to an “unsuitable” hybrid strategy. Not every patient and not every lesion are suitable for intraoperative stent placement, and careful consideration must be given to the appropriate indication for performing this procedure. Both percutaneous and intraoperative stent delivery offer very distinct advantages (Table 1), and the choice of the appropriate treatment strategy must be tailored to the individual patient. In addition, surgical pulmonary patch angioplasty is a valuable treatment alternative and may be more beneficial for certain lesions. This applies particularly to very calcified proximal branch PA stenoses. However, surgical therapy also provides distinct disadvantages compared to stent therapy that include difficulty in approaching distal lesions, the rate of restenosis, which is higher than after placement of endovascular stents, the risk of phrenic nerve injury and the invariable use of cardiopulmonary bypass, the latter being less of a problem if other lesions have to be corrected at the same time. In addition, stenotic lesions created by vascular folds or external compression are notoriously difficult to address surgically, even with a “perfect” patch repair. Other groups that may benefit from hybrid therapy include hemodynamically unstable patients who may require cardiopulmonary support, or patients in the early postoperative period who may require stent placement across freshly-operated lesions. Our series has shown that the majority of intraoperative stent placements are electively planned prior to the procedure. The decision to entertain a hybrid strategy must be made between the cardiac interventionalist and cardiothoracic surgeon, preferably at the time of previous transcatheter evaluation. In fact, it is a good practice to discuss any intended transcatheter PA rehabilitation in a patient who anticipates undergoing open-heart surgery in the near future with the cardiothoracic surgeon at the time of cardiac catheterization. That way, the most appropriate treatment strategy can be defined at a time when all treatment options are available to the patient. Close cooperation and discussion between the cardiothoracic surgeon and cardiac interventionalist are as essential in preparation for a hybrid procedure as during the procedure itself. Performing angiography or fluoroscopy in the OR can be a challenge and does require preprocedural considerations. Taking an ordinary C-arm to a standard OR may require the patient to be positioned in a more unconventional manner, so as not to interfere with the operating table itself. Most ORs have only limited space available, and this space is often further crowded by cardiopulmonary bypass and other equipment. Angiography in larger pediatric patients, even when performed with temporary occlusion of the opposite branch PA, usually requires power injection, thus a portable power injector should be readily available. The built-in calibration facilities of many of the C-arms are often suboptimal. It is of considerable benefit, if the patient has undergone preprocedural cardiac catheterization or multislice CT evaluation, to allow accurate measurements of the stenosis and adjacent PA, as well as the distance to the PA side branches. If preprocedural catheterization data are available, the decision of which stent and balloon to use should be made well in advance of entering the OR and the appropriate equipment should be readily available. The decision of whether to deploy the stents using direct vision, which invariably requires the use of cardiopulmonary bypass or angiographic/fluoroscopic guidance through direct vascular puncture on the beating heart, depends on the type of surgical procedure performed, the need for cardiopulmonary bypass, and whether preexisting landmarks are available from previous angiographic images. However, even if stents are deployed under direct vision, operators should have a low threshold for using (exit) angiography, as this may allow identifying any residual PA stenosis or other procedure-related problems. Stent malposition which occurred in 1 of our patients with PA VSD, could have been identified intraoperatively with exit angiography. Some patients, especially in the early postoperative period, may poorly tolerate percutaneous stent placement. This is frequently related to a stiff guidewire splinting open the tricuspid and/or pulmonary valve, made even worse by the presence of a long hemostatic sheath. Other factors that would favor a hybrid approach in the OR include extremely hypoplastic PA branches or disconnected PAs, which are associated with an increased risk of vascular injury during stent deployment. In these situations, surgical exposure in the OR may allow direct access to the vessel, without the use of cardiopulmonary bypass, while having bypass on standby if any complications occur. In very sick infants, this option may be preferable to performing a hybrid intervention under direct vision using cardiopulmonary bypass support. Lastly, access to a branch PA stenosis may be facilitated through direct MPA puncture in a much more direct fashion than using a percutaneous transcatheter route. While stent implantation without cardiopulmonary bypass under fluoroscopic and angiographic guidance is indicated and beneficial in some patients, the majority will undergo hybrid stent delivery performed under direct vision with the patient on cardiopulmonary bypass. The largest group of patients in our series are adults who underwent previous tetralogy of Fallot or PA VSD repair, pulmonary valve or conduit replacement and who had an associated isolated proximal PA stenosis or kink, which is similar to other reported series.6 This group of patients frequently poses a considerable challenge in the cardiac catheterization laboratory, with the desired stent diameter requiring very large sheaths, frequently exceeding the maximum size of braided or reinforced sheaths currently available in the U.S. The size of these patients and their branch PAs can make accurate stent positioning a cumbersome and time-consuming undertaking in the cardiac catheterization laboratory, whereas stent deployment under direct vision frequently can be performed within a small fraction of the time required for a percutaneous approach. When deploying stents under direct vision, endoscopy should be performed to inspect the lesion itself and to assess the distance to any of the PA side branches (Figure 8). For this purpose, the endoscope is advanced across the stenosis until the origin of a side branch can be visualized, which then allows the operator to measure the distance the endoscope has been advanced into the branch pulmonary artery (Figure 8). This provides a good estimate of the maximum stent length that should be used to avoid jailing of a PA side branch. The endoscope further aides in confirming accurate placement of the guidewire. Wire placement in a small PA side branch may lead to stent malposition, jailing of adjacent pulmonary vasculature, and possibly even vascular injury of the side branch during balloon expansion. For the same reason, operators should avoid the temptation to “quickly” deploy a stent without using a guidewire, as the position of the distal balloon tip cannot be reliably assessed during balloon inflation. Once the stent is delivered, the result is examined by endoscopy. Stent meshwork that slightly protrudes into the MPA can easily be crimped and folded to create a smoother surface and aid subsequent transcatheter interventions (Figure 5). Similarly, the OR offers the possibility to even shorten a stent if necessary.8 This allows delivery of a stent to exactly the desired length, while not having to worry that the sharper surfaces of the shortened stent will create a significant problem such as balloon puncture during stent deployment. In general, hybrid stent delivery is much more forgiving in terms of balloon rupture or stent migration during stent deployment than is transcatheter therapy, and a suboptimally deployed or not fully expanded stent can easily be removed under direct vision. Even stent delivery using direct vision with the patient on cardiopulmonary bypass can be further evaluated with exit angiography. This is usually best performed prior to taking the patient off cardiopulmonary bypass when there is low flow and the opposite branch PA is possibly clamped. The ideal setup for hybrid stent delivery in the OR involves the use of a specifically designed hybrid operating suite. In our own institutions, this setup has been available since November 2007 (Figure 6), and offers several advantages when compared to the use of a standard C-arm in a normal OR. These include a much larger OR, a fixed flat-panel image intensifier with superior image quality, accurate calibration and the ability to adjust angles more easily than would be possible with an ordinary C-arm. The hybrid OR also includes a power injector, as well as endoscopes and other equipment that is routinely used during hybrid therapies. While the table is longer than a standard operating table, there remains a limited amount of space for the orderly arrangement of all the wires and equipment that an interventional cardiologist is accustomed to. Therefore, OR staff and the interventional cardiologist must work closely and communicate well to avoid a wire accidentally “dropping” on the floor. In conclusion, hybrid stent delivery in the OR, using either direct vision or vascular puncture, is a suitable alternative to percutaneous stent delivery or surgical patch augmentation. Each technique offers specific advantages and disadvantages, and as such, the ultimate approach chosen must be tailored to the individual patient. Using a well-equipped hybrid operating suite facilitates safe intraoperative stent delivery in a wide variety of patients. Close cooperation between surgical and interventional teams is essential for successful outcomes.

1. Ing FF, Grifka RG, Nihill MR, Mullins CE. Repeat dilation of intravascular stents in congenital heart defects. Circulation 1995;92:893–897.
2. Mullins CE, O’Laughlin MP, Vick GW III, et al. Implantation of balloon-expandable intravascular grafts by catheterization in pulmonary arteries and systemic veins. Circulation 1988;77:188–199.
3. O'Laughlin MP, Perry SB, Lock JE, Mullins CE. Use of endovascular stents in congenital heart disease. Circulation 1991;83:1923–1939.
4. Bacha EA, Marshall AC, McElhinney DB, del Nido PJ. Expanding the hybrid concept in congenital heart surgery. Seminars in Thoracic & Cardiovascular Surgery 2007;Pediatric:Cardiac-50.
5. Hjortdal VE, Redington AN, de Leval MR, Tsang VT. Hybrid approaches to complex congenital cardiac surgery. Eur J Cardiothorac Surg 2002;22:885–890.
6. Mitropoulos FA, Laks H, Kapadia N, et al. Intraoperative pulmonary artery stenting: An alternative technique for the management of pulmonary artery stenosis. Ann Thorac Surg 2007;84:1338–1341.
7. Zartner P, Cesnjevar R, Singer H, Weyand M. First successful implantation of a biodegradable metal stent into the left pulmonary artery of a preterm baby. Catheter Cardiovasc Interv 2005;66:590–594.
8. Ing FF. Delivery of stents to target lesions: Techniques of intraoperative stent implantation and intraoperative angiograms. Ped Cardiol 2005;26:260–266.
9. Bokenkamp R, Blom NA, De Wolf D, et al. Intraoperative stenting of pulmonary arteries. Eur J Cardiothorac Surg 2005;27:544–547.