We describe a novel, cost-effective, and efficient technique for percutaneous delivery of three-dimensionally printed coronary implants to create closed-chest swine models of ischemic heart disease. The implants were fixed in place using a mother-and-child extension catheter with high success rate.
Minimally invasive methods for creating models of focal coronary narrowing in large animals are challenging. Rapid prototyping using three-dimensionally (3D) printed coronary implants can be employed to percutaneously create a focal coronary stenosis. However, reliable delivery of the implants can be difficult without the use of ancillary equipment. We describe the use of a mother-and-child coronary guide catheter for stabilization of the implant and for effective delivery of the 3D printed implant to any desired location along the length of the coronary vessel. The focal coronary narrowing was confirmed under coronary cineangiography and the functional significance of the coronary stenosis was assessed using gadolinium-enhanced first-pass cardiac perfusion MRI. We showed that reliable delivery of 3D printed coronary implants in swine models (n = 11) of ischemic heart disease can be achieved through repurposing mother-and-child coronary guide catheters. Our technique simplifies the percutaneous delivery of coronary implants to create closed-chest swine models of focal coronary artery stenosis and can be performed expeditiously, with a low procedural failure rate.
Ischemic heart disease continues to be the number one cause of death in the United States1. Large animal models have been used experimentally to understand and characterize mechanisms driving coronary artery disease (CAD) and associated complications (including myocardial infarction, arrhythmic events, and heart failure), as well as for testing of new therapeutics or diagnostic modalities. Results from these studies have helped to broaden the understanding, diagnosis, and monitoring of ischemic heart disease and to advance clinical practice2. Several animal models including rabbits, dogs, and swine have been used. However, coronary stenoses, particularly discrete lesions, occur very rarely in these animals and are difficult to induce reproducibly3. Prior work described the creation of artificial coronary stenoses using ligation, occluders, or external clamps. Recently, we described how to use 3D printing technology to manufacture coronary implants that can be used to percutaneously create discrete artificial coronary narrowing4. Using computer-aided design software, we designed coronary artery implants as hollow tubes with varying inner and outer diameters as well as implant length and then fabricated them using commercially available additive materials. The implants are smooth, hollow, 3D printed tubes with rounded edges. We designed a library of implant sizes with a range of inner diameter, outer diameter, and length. The outer diameter of the implant is based on the size of the coronary guide catheter. The inner diameter is based on the size of a deflated coronary angioplasty balloon. We varied the length of the implant to tailor the desired severity of perfusion. However, safe percutaneous delivery of such devices can be challenging due to the lack of wires and catheters manufactured specifically for large animal use. In contrast, an extensive collection of catheters, wires, and supportive equipment are available for clinical use in human coronary arteries. In this work, we show how to repurpose a clinical grade mother-and-child coronary guide catheter for the delivery of the 3D printed coronary implants.
The GuideLiner catheter (Figure 1A) was developed for percutaneous coronary intervention (PCI) to allow for deep catheter seating and increased support for complex cases5. In our investigation, the GuideLiner catheter was chosen due to familiarity of use and availability, but similar catheters, where available, may also be considered. Considered a "mother-and-child" guide catheter (Figure 1B), the device fits inside a typical coronary guide catheter ("mother") and is a coaxial flexible tube ("child"). This catheter can be inserted over a guidewire and effectively lengthens the reach of a typical coronary guide catheter by extending beyond the end of the coronary guide. The GuideLiner or a similar mother-and-child catheter can be used as added support for deployment of the 3D printed coronary implants. Because the implants are mounted over angioplasty balloons to be inserted as a unit over a coronary wire into the vessel (Figure 1B,1C), the catheter offers additional support to deliver the implant to the desired site. By positioning the mother-and-child catheter just proximal to the balloon, the implant remains at the desired location during balloon deflation and retraction. Despite having some firmness to its structure, the mother-and-child catheter's unique ability to be advanced deep into coronary arteries over a guidewire and the radiopaque marker at the catheter tip were essential characteristics for implantation.
Our assembled delivery apparatus consisted of a typical coronary guide catheter, the mother-and-child catheter, and a 3D printed implant fixed onto a deflated coronary angioplasty balloon (Figure 1B). As a functional delivery unit, the mother-and-child catheter not only provided stable additional support for the delivery of the equipment but was also uniquely applied as a shearing device to keep the implants in place during deflation and removal of the balloon. The radiopaque marker at the catheter tip served as a positioning guide for the assembled apparatus and sits proximal to the angioplasty balloon. These characteristics allowed for precise deployment of the flow-limiting implants. The process was designed to be reproducible, efficient, and humane for the animal subjects.
In our application, the mother-and-child percutaneous delivery technique was used to create swine models with focal coronary stenosis for evaluation of contrast-enhanced stress cardiac perfusion magnetic resonance imaging (MRI). However, the technique may be employed in other investigations including vascular systems outside the coronary vessels.
We conducted the experiments according to the guidelines by the Animal Welfare Act, the National Institutes of Health, and the American Heart Association on Research Animal Use. Our Institutional Animal Care and Use Committee approved the animal study protocol.
1. Preprocedural preparation of 3D printed coronary stenosis implants
2. Preprocedural preparation of animal subjects
3. Procedural anesthesia
4. Vascular access
5. Preprocedural medication administration
6. Hemodynamic monitoring
7. Preparation of implant delivery equipment
8. Coronary angiography and deployment of coronary implant
After initial optimization of the procedure, the intervention component was completed within 30 min. The implants were successfully delivered in all 11 subjects (100%). The implant was retrieved at the autopsy in all 11 subjects (100%). Using the diagonal branches (along the LAD) or obtuse marginal branches (along the LCX) as positional markers, we found the position of the implant at fluoroscopic-guided deployment and at autopsy to be consistent in 10 of the 11 (91%) subjects where the implant was retrievable. In one subject, there was slight distal migration of the implant, which may be related to vasodilation induced by intracoronary nitroglycerin injection for coronary spasm. Of the 11 subjects studied, 9 survived for the entire catheterization and completed the MRI protocol, giving us an 82% procedural success rate. Two subjects died after the implants were deployed. The first subject developed ventricular fibrillation in the MRI suite well after deployment of the implant. The second died in the MRI scanner in the setting of hypotension midway through the experiment. At the time of dissection, we did not see thrombus within the implants or other signs of structural injury to the vessels. The high survival rate (2 deaths, 9 of 11 survived) highlights the importance an effective anti-arrhythmic prophylaxis regimen. An illustrative example of stress cardiac perfusion MRI is provided in Figure 6. Detailed implant design and full results of the MRI validation will be reported separately.
Figure 1: Catheter design and assembled apparatus with mounted coronary implant. (A) Diagram of the components of the mother-and-child catheter6. (B) Assembled apparatus showing the coronary balloon inflated with the 3D printed implant mounted and fixed at the leading head of the catheter, which protrudes through the guide catheter. (C) A magnified image of the 3D printed implant is shown mounted onto the angioplasty balloon. Please click here to view a larger version of this figure.
Figure 2: Coronary angiogram in the anteroposterior projection shows selective contrast-enhancement of the left main coronary artery system. Please click here to view a larger version of this figure.
Figure 3: Coronary angiogram in the anteroposterior projection shows the 0.014" 300 cm coronary wire in the left anterior descending artery. Please click here to view a larger version of this figure.
Figure 4: Coronary angiogram in the anteroposterior projection. The image on the left shows the assembled mother-and-child catheter with the inflated coronary balloon and implant in the mid to distal segment of the left anterior descending artery. A higher magnification of the assembled apparatus within the coronary vessel is shown in the right panel. Please click here to view a larger version of this figure.
Figure 5: Anteroposterior angiogram. The image on the left shows a focal stenosis in the distal left anterior descending artery after deployment of the implant. A higher magnification of the discrete coronary narrowing induced by the implant is shown in the right panel. Please click here to view a larger version of this figure.
Figure 6: Stress cardiac perfusion magnetic resonance images of a coronary implant deployed in the proximal to mid left anterior descending artery. The images at rest (upper panel) and peak adenosine vasodilator stress (lower panel) show inducible perfusion defects in the segments subtended by the left anterior descending artery. Please click here to view a larger version of this figure.
Figure 7: Autopsy images. (A) The implant at the distal left anterior, descending artery. (B) The absence of gross injury to the coronary vessel. (C) Implant without thrombus. Please click here to view a larger version of this figure.
Figure 8: Histopathology of swine myocardial tissue. (A) Gross pathology and (B) triphenyltetrazolium chloride stains in one subject showed no evidence of myocardial tissue infarction. Please click here to view a larger version of this figure.
In this work, we focused on a novel percutaneous deployment strategy for coronary stenosis-inducing implants and showed that a mother-and-child catheter can be repurposed for effective percutaneous delivery of 3D printed coronary implants. Discrete artificial coronary stenoses of variable severity can be created quickly in swine models with a high success rate and in a minimally invasive manner using standard human percutaneous coronary interventional techniques and equipment. These implants were shown to be safe in the acute setting and were also effective at creating severe angiographic stenoses, which correlated with stress-induced perfusion defects during vasodilator stress cardiac MRI. Compared to open-chest techniques, percutaneous delivery of stenosis-inducing implants is less invasive and more humane.
There are several other minimally invasive techniques currently available to create flow reduction in large animal models. The 3D printed coronary implants differ fundamentally from balloon occlusion and coil occlusion in that the stenoses induced by the 3D printed implants do not completely occlude the vessel. This is a major difference that allows for modelling of stress-induced ischemia rather than infarction7,8. Rissanen et al.9 describe a percutaneous technique that creates flow limiting, non-obstructive stenoses in swine models using a coronary stent wrapped in a polytetrafluoroethylene tubing. The tubing could be shaped by employing needles and heat to create luminal narrowing of various degrees. It is clear that the implants we used differ in design and thorough description with full validation is beyond the scope of the current work, which is to describe the novel methodology used for delivery of 3D printed coronary implants. Utilizing the mother-and-child catheter allowed for precise deployment of the implants deep in the coronary arteries. It is difficult to compare procedural success between our studies as other investigators explored a chronic model and kept the swine alive for an extended period of time9. Bamberg et al. described a method using balloon catheters inflated within 3 mm stents to create stenoses of 50% and 75% in the left anterior descending artery. This latter method differs from our investigation in that the stenoses created required catheters to be left inside the animals. There is no way to create an artificial lesion and remove all equipment. While viable, the Bamberg method does not allow for investigation of ischemia beyond the acute setting and residual wires would cause image artifacts10.
The role of mother-and-child catheters in coronary interventions has been well established, but their use to deliver implants into vascular beds has not been previously described5,6. The two most challenging aspects of percutaneous implant delivery include selective deployment into a precise coronary segment and prevention of retrograde migration. Attempting to deploy the device over angioplasty balloons was not effective because the implant could be pulled proximally in the vessel after balloon deflation. For several reasons, the mother-and-child catheter proved to be a valuable tool for fixing the implants in place during balloon withdrawal. The mother-and-child catheters fit easily in the coronary guide catheters and their size was ideal for our intervention. They were slightly larger than the deflated coronary balloon, allowing us to shear the implant off and to prevent retrograde migration of the implant as the balloon was withdrawn. The support provided by the mother-and-child catheter enabled the implants to be deeply seated in the coronary artery with strong apposition to the vessel lumen. Additionally, the radiopaque marker on the tip of the mother-and-child catheter helped position the catheter just proximal to the implant, as identified by the marker on the delivery balloon. Though the technique was mostly effective, in one subject there was slight distal migration after implant delivery. This may have been due to injection of intracoronary nitroglycerin for coronary vasospasm and resultant vasodilation leading to distal migration of the implant. The GuideLiner catheter was chosen due to familiarity of use, but there are a number of other similar devices which could potentially be used in its place. The Guidezilla Guide Extension Catheter (Boston Scientific, Marlborough, Massachusetts, USA) is also available in a 6F size and has a similar structure to the GuideLiner. There is also a Guidion rapid exchange guide extension catheter (Interventional Medical Device Solutions, Roden, The Netherlands) which comes in sizes 5–8F and could also potentially be used in place of the GuideLiner catheter.
Our deployment technique can be performed efficiently and humanely in swine with a low procedural failure rate. In our preliminary study the procedural failure rate was 18%. There was a learning curve associated with the technique as we streamlined our interventions. However, despite the learning curve, all animal subjects survived the initial implant deployment intervention. The lesions created were focal and the narrowing ranged in severity, but they were not occlusive. These stenoses were angiographically significant and produced inducible perfusion defects during stress perfusion MRI. Figure 6 is an example of a focal perfusion defect seen on MRI after successful implant deployment to the LAD. We aimed to create ischemia rather than infarction. Figure 8 shows an example of histopathologic analysis of the myocardial tissue, which shows no evidence of infarction. The method relies on human coronary angioplasty equipment, and the similarity in swine coronary size to those of humans. The outer diameter of the 3D printed implant was based on the inner diameter of the guiding catheter and the inner diameter of the mother-and-child catheter. The minimal luminal diameter of the stenosis was based on the size of the deflated coronary balloon. The final flow-limiting severity of the discrete stenosis is based on the inner diameter and the length of the implant. Although resting angiographic flow was preserved, maximal coronary blood flow was reduced, as evidenced by the MRI perfusion scans. Future work will focus on replacing the balloon delivery wire with a pressure wire and measurement of fractional flow reserve or instantaneous flow reserve. Similarly, downstream microvascular injury can be produced by local injections of microspheres either through the delivery balloon or the mother-and-child catheter itself.
Our low procedural failure rate in a closed-chest swine model shows promise for future implementation. Because complete total occlusion was not performed, myocardial infarction was avoided, and may have contributed to the lower rate of malignant arrhythmias. In our study only 1 subject developed ventricular fibrillation. After an initial period of optimization, we cut down procedural time to roughly 30 min per case.
In summary, our results demonstrate a novel technique for deployment of 3D printed coronary implants and show the feasibility of creating a closed-chest swine model of discrete focal coronary stenosis. This minimally invasive technique can be used for testing and development of new diagnostic imaging techniques in ischemic heart disease. We used stress cardiac perfusion MRI, but other modalities may include nuclear imaging, ultrasound, and computed tomography. Although this model is immediately applicable to ischemic heart disease, with minor modifications, the technique can be employed for other occlusive vascular disease states.
The authors have nothing to disclose.
We thank staff members at the UCLA Translational Research Imaging Center and the Department of Laboratory Animal Medicine at University of California, Los Angeles, CA, USA for their assistance. This work is supported in part by the Department of Radiology and Medicine at David Geffen School of Medicine at UCLA, the American Heart Association (18TPA34170049), and by the Clinical Science Research, Development Council of the Veterans Health Administration (VA-MERIT I01CX001901).
3D-Printed coronary implants | Study Site Manufactured | ||
Amiodarone IV solution | Study Site Pharmacy | ||
Amplatz Left-2 (AL-2) guide catheter (8F) | Boston Scientific, Marlborough, Massachusetts, USA | ||
Balance Middleweight coronary wire (0.014" 300cm) | Abbott Laboratories, Abbott Park, Illinois, USA | ||
COPILOT Bleedback Control valve | Abbott Laboratories, Abbott Park, Illinois, USA | ||
Esmolol IV solution (1 mg/kg) | Study Site Pharmacy | ||
Formlabs Form 2 3D-printer with a minimum XY feature size of 150 µm | Formlabs Inc., Somerville, Massachusetts, USA | ||
Formlabs Grey Resin (implant material) | Formlabs Inc., Somerville, Massachusetts, USA | ||
Gadobutrol 0.1 mmol/kg | Gadvist, Bayer Pharmaceuticals, Wayne, NJ | ||
GuideLiner catheter (6F) | Vascular Solutions Inc., Minneapolis, Minnesota, USA | ||
Heparin IV solution | Surface Solutions Laboratories Inc., Carlisle, Massachusetts, USA | ||
Ketamine IM solution (10 mg/kg) | Study Site Pharmacy | ||
Lidocaine IV solution | Study Site Pharmacy | ||
Male Yorkshire swine (30-45 kg) | SNS Farms | ||
Midazolam IV solution | Study Site Pharmacy | ||
NC Trek over-the-wire coronary balloon | Abbott Laboratories, Abbott Park, Illinois, USA | ||
Oxygen-isoflurane 1-2% inhaled mixture | Study Site Pharmacy | ||
Rocuronium IV solution | Study Site Pharmacy | ||
Sodium Pentobarbital IV solution (100mg/kg) | Study Site Pharmacy | ||
Triphenyltetrazolium chloride stain | Institution Pathology Lab |