Flexible 3D printed models of double outlet right ventricle for hands-on training of suturing the ventricular septal defect patch (photo courtesy of Prof. Shi-Joon Yoo from The Hospital for Sick Children, University of Toronto, Toronto, Canada)
The benefits of 3D printing are under-realized among the septal defects. 3D printing from CT, MRI, and 3D echocardiography, and the post-processing steps used to generate these models, is highly valuable and amendable to design patches and to navigate the direction of intraoperative occluder devices (Chaowu et al. 2016; Kim et al. 2008). 3D printing of the heart can also include the great vessels (Riesenkampff et al. 2009; Schmauss et al. 2015; Shiraishi et al. 2014, 2010; Noecker et al. 2006; Olivieri et al. 2015; Giannopoulos et al. 2015; Samuel et al. 2015). Other examples include ostium secundum ASD (Faganello et al. 2015), preoperative evaluation of ASD, and occlusion trials to avoid potentially unnecessary procedures (Chaowu et al. 2016) and VSDs (Olivieri et al. 2015). 3D printing has been used for occluder device sizing and selection of the approach to cross the defect in a congenital muscular VSD (Kim et al. 2008).
7.2.1 Complex Pediatric and Adult Congenital Heart Diseases
Reports of 3D printing benefits in congenital heart disease are now too numerous to discuss in a case-by-case format (Riesenkampff et al. 2009; Noecker et al. 2006; Matsumoto et al. 2015; Mottl-Link et al. 2008; Olivieri et al. 2014; Ryan et al. 2015; Shirakawa et al. 2016; Sodian et al. 2008; Valverde et al. 2015). The most important contribution in the field is in double outlet right ventricle (DORV) (Farooqi et al. 2016; Yoo et al. 2016a, b; Greil et al. 2007; Vodiskar et al. 2017). In addition to the variability associated with the VSD (usually present), the infundibular and intracardiac variability has resulted in individualized surgical approaches. These have been summarized and captured in a library (IMIB-CHD n.d.) of flexible 3D printed models used to teach anatomy and the aid in several organized surgical training initiatives (Yoo et al. 2016a, b).
Tetralogy of Fallot (TOF) also has variability, and infants have greatly benefited from the availability of 3D models. Examples include TOF with pulmonary atresia, where 3D printing has been used to depict the pulmonary vascular anatomy, including collateral flow (Ryan et al. 2015) that can be referenced intraoperatively (Ngan et al. 2006). Benefits have also been realized in infants with hypoplastic left heart syndrome (Shiraishi et al. 2006, 2014; Kiraly et al. 2016) and transposition of the great vessels (Valverde et al. 2015).
7.3 Adult Heart Disease
7.3.1 Left Atrial Appendage Closure
There is an increasing need to exclude circulation within the left atrial appendage for patients with non-valvular atrial fibrillation; this can be an alternative therapy to prevent thromboembolism when there is a relative contraindication to anticoagulation (Holmes et al. 2014). The current standard of care uses TEE with fluoroscopy guidance, and there is an increasing interest to utilize data from a pre-intervention CT, akin to other methods for atrial fibrillation therapy. The left atrial appendage has a variable anatomy, and while there are now several devices available, there can be considerable debate and uncertainty regarding the optimum device sizing. Once the decision to use a device is made clinically, sizing is of paramount importance, since there can be serious complications to a procedure that leads to incomplete occlusion or to one that uses a device whose size exceeds the tolerance of the tissue. The role of 3D printing is to assist in best-sizing the device to determine the optimum dimensions and enabling simulation and education of the procedure. The latter application will benefit from new printing materials to better emulate myocardium.
7.3.2 Hypertrophic Obstructive Cardiomyopathy
Hypertrophic obstructive cardiomyopathy has a wide spectrum of disease states but in general has the pathophysiology of eccentric and regional hypertrophy of the left ventricle (Maron et al. 2016). While most patients are treated medically, myomectomy of the ventricular septum can be performed with the intended result of reducing the obstruction to the outflow of blood to the aorta (Gersh et al. 2011; Elliott et al. 2014; Maron et al. 2011). Imaging is usually performed with cardiac MRI. 3D models offer a unique perspective of the 3D orientation of the outflow tract and key haptic anatomical feedback (Yang et al. 2015) (Fig. 7.2). In theory, a 3D model that could incorporate the systolic anterior motion from the mitral valve anterior leaflet would be highly valuable as it could be connected to a flow pump to spatially comprehend the motion and its relationship to the outflow tract throughout the cardiac cycle.
Top left: Hypertrophic interventricular septum (asterisks), posterior papillary muscle (P), and intraventricular muscle band or accessory papillary muscles (arrowhead). Top right: Bull’s eye map from end-diastolic CT demonstrating extent of hypertrophy (red area, >15 mm in thickness). Middle row: Color-coded 3D-printed model demonstrates the hypertrophic septum (asterisks), papillary muscle (A anterior, P posterior), and intraventricular muscle band (asterisks). Bottom row: Intraoperative view from the apical approach demonstrates the limited visual field of the LV cavity in both the model and the patient. AO aorta, LA left atrium, LV left ventricle, MV mitral valve. Yang et al. Circulation. 2015;132:300–301
7.3.3 Cardiac Tumors
3D printing provides advanced understanding of the relationship of the unusual cardiovascular tumors for which intervention is a consideration. While 3D visualization, e.g., using customized MR acquisitions, can be adequate for delineating the myocardium, models can be useful to show the relationship between a mass and an adjacent structure that may be involved (Jacobs et al. 2008; Schmauss et al. 2013; Son et al. 2015; Al Jabbari et al. 2016).
7.3.4 Valve Disease
The last decade has seen highly innovative treatment strategies for valve disease. Two major procedures that have been enabled by technique and device advances are transcatheter aortic valve repair (TAVR) (Schmauss et al. 2012; Webb and Lauck 2016) and transcatheter mitral valve repair (TMVR) . These two procedures share two main characteristics in common. First, they are far less invasive than conventional valve treatments that use open surgery. However, the second characteristic is a consequence of the first: without surgery, there is no opportunity to visually inspect and understand the true 3D representation of the valve anatomy. This can be highly relevant as the geometry differs among patients, and the experience of the operator in these relatively new procedures is variable. Consequently, 3D printing of valve pathology has emerged as a growing field, taking advantage of the ability of 3D printing to incorporate multimodality imaging, namely, echocardiography plus CT and MRI.
The number and scope of TAVR continues to grow (Webb and Lauck 2016), and it is now established that TAVR is a generally safe alternative to surgery for many patients (Nishimura et al. 2014; Moat 2016) and has an expanding role (Webb and Lauck 2016). While some procedures are straightforward, there is mounting evidence that 3D printing has a clear role to help delineate both the anatomy and the hemodynamics of pathology, as well as the effect of calcification for patients at higher risk for complications (Gallo et al. 2016). Valvular stenosis of the aorta is already well addressed by 3D printed models, and these models can be used to plan TAVR (Maragiannis et al. 2015) (Fig. 7.3). In some patients, the procedure is designed through the cardiac apex, where a printed model of the myocardium in addition to the valve can help plan intervention (Fujita et al. 2015). There is also an increasing interest in so-called “valve-in-valve procedures” where a second valve is placed within the first, generating another potential indication where a physical model can be used to classify patients into surgical versus percutaneous candidate, and when a second TAVR is considered, to determine the most accurate measurements that will lead to optimal choice of prostheses (Fujita et al. 2015).
3D Printing of aortic stenosis . 3D Printed model of a severely stenotic aortic valve derived from CT (a, b); aortic wall tissue is printed in flexible transparent material and calcium in opaque rigid material. Flow experiments using this model resulted in functional characteristics similar to those of in vivo assessed by spectral Doppler. Transcatheter-deployed valve in a 3D printed model (c) seen from endoscopic LVOT (d) and aortic views (e) demonstrate the final configuration of a self expanding stent around a calcification (red asterisk in e). Maragiannis D et al. J Am Coll Cardiol. 2014 Sep 9;64(10):1066–8 (panels a–c). Maragiannis D et al. Circ Cardiovasc Imaging. 2015;8:e003626 (panels d, e)
The implementation of more flexible materials, and those that better mimic physiology in health and disease, is an important advance so that models can better emulate the impaired function of the stenosed aortic valve (Maragiannis et al. 2014). Regarding complications, there is ongoing debate regarding the significance of small paravalvular leaks. However, when a leak is characterized as moderate, there is a negative impact on valve function, as well as survival (Figulla et al. 2016). When intervention for a leak is considered the best management option, a percutaneous approach is often preferred (Sorajja et al. 2011), where models can help guide the procedure. Conversely, flexible 3D printed models of the aortic root complex derived from routine pre-TAVR CT (Dill et al. 2013) have been shown to predict paravalvular leaks after the procedure as determined by echocardiography (Ripley et al. 2016), which may help minimize this complication.
The success of less invasive interventions in the aortic valve has undoubtedly inspired approaches to treating mitral disease. Several studies support that the valve apparatus can be printed (Binder et al. 2000; Dankowski et al. 2014; Kapur and Garg 2014; Mahmood et al. 2014, 2015; Witschey et al. 2014). The imaging data required for these models is derived from echocardiography and CT, and there is growing evidence that this as an appropriate indication for 3D printing. For example, prints of normal versus regurgitant mitral valves (Witschey et al. 2014) have been used to determine the ring selection at annuloplasty and have been used to enhance spatial understanding of the 3D relationships at surgery (Owais et al. 2014) and estimate the risk of left ventricle outflow tract obstruction (Wang et al. 2016).
Regarding transcatheter approaches, these percutaneous techniques, in theory, are lower risk than their counterpart open procedure for functional mitral regurgitation (Figulla et al. 2016). As in TAVR, though, the lower risk from avoiding an open repair comes with the cost of reduced intra-procedure visualization, and there is an unmet need to correctly size the mitral annulus and to avoid obstruction of left ventricular outflow. Literature is beginning to support models for pre-percutaneous implantation of an annuloplasty system (Dankowski et al. 2014) as well as deployment of a MitraClip (Abbott Laboratories, Abbott Park, IL) (Little et al. 2016) (Fig. 7.4).
Simulation of patient-specific mitral valve intervention using 3D printed model from CT. Model valve leaflets and the subvalvular calcium deposition (upper left panel) was created from CT images to assist in selection and sizing of an occluder device in a case of severe mitral valve regurgitation with restricted leaflet coaptation and a perforation of the posterior leaflet (upper right panel). An AMPLATZER Duct Occluder II (St. Jude Medical, St. Paul, Minnesota) was placed across the posterior leaflet perforation (lower left panel) and evaluated for potential interaction with the leaflet coaptation zone (lower right panel; superimposed dotted line). Little SH et al. JACC Cardiovasc Interv. 2016 May 9;9(9):973–5
7.4 3D Printing for the Systemic Vessels
Large vessels are readily 3D printed using flexible materials that are amendable for printing a hollow lumen with, e.g., aneurysms, mobile thrombi, and atherosclerotic plaques (Fig. 7.5); benefits span many vascular pathologies (Pepper et al. 2013; Tam et al. 2014; Itagaki 2015; Salloum et al. 2016). In addition, flow simulations using these models, printed with appropriate methods to replicate vascular compliance (Biglino et al. 2013), expand the utility of these models beyond assessing morphology alone. Common applications include aneurysms, for example, on the root of the aorta in patients with Marfan syndrome. Models enable patient-tailored approaches such as customized patch design for repair of the aorta (Izgi et al. 2015). This can minimize risk and preserve as much of the native aorta as possible (Treasure et al. 2014; Tam et al. 2013). Another application in the aorta is the use of 3D printing to improve outcomes in endovascular aneurysm repair, where a physical model can accurately depict complex aneurysm geometry (Tam et al. 2014; Russ et al. 2015).
CTA of aorta with mobile mural thrombi (left panel, green arrows) and calcifications (left panel, white arrowheads) is used to 3D-print a model that includes a cutout window that can be removed to inspect the aortic lumen to appreciate the location and size of calcifications and thrombi toward planning percutaneous intervention. Giannopoulos AA et al. J Thorac Imaging. 2016 Sep;31 (5):253–72
Finally, 3D printing provides a uniquely strong methodology to model and test cardiovascular hemodynamics. Printed vascular models provide the opportunity for patient-specific device bench testing (Russ et al. 2015; Meess et al. 2017) and even optimization of imaging technologies and hypothesis testing that would not otherwise be possible in vivo (Mitsouras et al. 2015; Nagesh et al. 2017).
3D printing is rapidly being developed and its applications are expanding in the cardiovascular arena. Models are now routinely generated from CT and MRI images, and increasingly from 3D echocardiography. The primary use of models to date has been in surgical planning, although there is a growing interest to use models for intravascular procedure planning, outcome prediction, and even diagnosis for complicated patients. Benefits include patients with CHD, valve diseases, and in particular those amenable for newer, less invasive treatments, certain forms of structural heart disease, and for vascular pathologies particularly in the aorta. This new “modality” represents a paradigm shift from the last two decades in which 3D visualization on a 3D screen changed the way that pathology was depicted. Over the next decade to 20 years, models will be generated directly from noninvasive imaging, further simplifying management of even the most complicated patients and providing new opportunities for care pathways.