3D Printing for Left Atrial Appendage (LAA) Modeling Based on Transesophageal Echocardiography: A Step Forward in Closure with LAA Devices

3D printing is the process of creating 3-dimensional objects by the integration of digitized images. The dedicated printer arranges layers of a material (plastic, resin or metal) in a volumetric process in order to replicate the interrogated object. In the case of the heart, an imaging modality is chosen, either computed tomography (CT) or magnetic resonance imaging (MRI) and, more recently, 3D echocardiography datasets. A volumetric dataset is obtained and following the respective anatomical segmentation, the model can be smoothed out except for noise.Potential cardiac applications are the comprehensive appraisal of complex congenital anomalies for dedicated surgical corrections as well as the detailed description of the aortic valve root anatomy for transcatheter aortic valve implantation procedures and left atrial appendage (LAA) anatomy for closure device applications [1].In a recent paper on 2 cases for LAA occlusion, it was demonstrated that the 3D model could help to find the correct position within the LAA, size the device and guide the choice of the closure device regardless of the measurements provided by angiography and transesophageal echocardiography (TEE) [2].The paper by Liu et al. [3 ]in this issue of Cardiology provides a series of 8 patients with 3D printing models of LAA produced from 3D datasets derived by 3D TEE. LAA modeling was used for both the preinterventional reference and the procedural simulation of LAA closure. Moreover, device sizing was tested to eliminate anatomic discrepancies. These efforts represent the frontier for applied personalized medicine in the field of LAA intervention.Atrial fibrillation (AF) occurs in approximately 0.4-1% of the general population, increasing with age to >8% in those >80 years of age, with the prevalence expected to have doubled during the next 20 years [4,5,6]Focused interrogation should be applied to the LAA when evaluating patients with AF to determine the risk for cardioembolic complications, especially before proceeding with cardioversion.Stroke represents a key cause of morbidity and mortality from cardiovascular disease, with an annual incidence of almost 800,000 patients with a new or recurrent stroke and an estimated prevalence of 7 million patients in the USA statistics; almost 80% of strokes are caused by focal cerebral ischemia due to arterial occlusion [4]. The incidence of strokes due to AF increases steeply from 1.5% at 50-59 years of age to >20% in octogenarians [7]. Strokes related to AF are associated with a higher mortality and morbidity when compared with non-AF strokes, emphasizing the need for effective stroke prevention in AF [8].The LAA stems from the primordial left atrium (LA), which is formed by the adsorption of the primordial pulmonary veins and their branches [9].There are variations in LAA size, shape and the relationship with adjacent cardiac and extracardiac structures, all of which can be extremely important when interventional procedures are performed.In postmortem hearts, the LAA usually has 2 lobes (in 54%) and less common are 3 lobes (in 23%), 1 lobe (in 20%) and 4 lobes (in 3%), and no significant age- or sex-related differences in LAA morphologies have been noted [10]. An increased number of lobes is associated with the presence of a thrombus independent of other predisposing factors [11].Using multidetector CT (MDCT) and cardiac MRI, the shapes of the LAA in AF have been classified into 4 morphological types, with the ‘chicken wing' being the most common (48%), followed by ‘cactus' (30%), ‘windsock' (19%) and ‘cauliflower' (3%). The ‘cauliflower' morphology is most often associated with an embolic event [12].Casts of the inner surface of the LAA reveal complex indentations made by the pectinate muscles that line the cavity of the appendage.The LAA is the site in the LA where >90% of thrombi are found in nonvalvular AF in transesophageal studies [13]. Thrombus formation is more likely to occur within the LAA when reduced contractility and stasis ensue, as in the presence of AF, with a decrease in Doppler velocities and dilation of the LAA.Patients with significant left ventricular dysfunction and elevated left ventricular end-diastolic pressure may also be at risk for LAA thrombus formation despite the absence of AF. LAA thrombi may be present in 15-20% of patients with dilated cardiomyopathy who are in sinus rhythm [14]. LAA thrombi are present in up to 14% of patients with acute (<3 days) AF [15]. Moreover, thrombus formation may develop even in patients with AF who are receiving therapeutic anticoagulation therapy. On TEE, it was found that up to 1.6% of patients treated with anticoagulation for 1 month had evidence of an LAA thrombus [16].After cardioversion, temporary stunning, documented as a reduction in LAA flow velocities, and a worsening mechanical function of both the LA and the LAA, can appear, predisposing to thrombus formation [17]. LAA stunning after cardioversion occurs whether conversion to sinus rhythm is spontaneous [18] or is associated with direct current cardioversion (either external [19] or low-dose internal [20]) or pharmacological cardioversion [21] and it may be modified by irbesartan pretreatment [22].TEE imaging of the LAA is best obtained using a multiplane approach in both long- and short-axis views as well as with 3D imaging. Echo contrast eliminates many of the artifacts and provides complete opacification of the LAA in the absence of filling defects due to thrombus [23]. Despite the evidence that patients with dense spontaneous echo contrast within the LAA have a stroke rate of 18.2% per year if not treated with warfarin and a 4.5% per year with warfarin treatment, the presence of an LAA thrombus itself triples the overall stroke rate [24].Although 2D TEE provides higher-resolution images due to the better frame rate, 3D TEE allows a more comprehensive anatomic assessment of the LAA. However, there are only limited data about 3D TEE accuracy for detecting LAA thrombi. LAA device closure is an evolving treatment strategy to prevent embolic events in patients with nonvalvular AF [25]. There are currently 2 strategies available for percutaneous LAA closure: LAA occlusion and LAA exclusion. Occlusion is defined as placement of a device into the LAA percutaneously, and exclusion refers to the application of an external ligature to isolate the LAA from the circulation (LARIAT suture delivery device, SentreHEART Inc., Redwood City, Calif., USA). The commonly used devices for LAA occlusion are the WATCHMAN device (Boston Scientific Corp., Natick, Mass., USA) and the AMPLATZER Cardiac Plug (ACP, St. Jude Medical, Inc., St. Paul, Minn., USA). Noninvasive imaging modalities, including 2D and 3D TEE and MDCT, are used for the assessment of overall LAA anatomy, orifice geometry and ‘landing zone' diameters, in order to safeguard the implantation of the device. MDCT provides valuable information about the shape of the LAA and for defining its relationships with surrounding critical structures, such as the left upper pulmonary vein and the left circumflex artery. Fusion-integrated imaging would increase the confidence for anatomic accuracy, assist in guidance and increase procedural efficiency [26].In Europe, the WATCHMAN device and the ACP are at present already widely used, in particular in patients with nonvalvular AF who have an absolute or relative contraindication to anticoagulation.Therapy with warfarin or the novel anticoagulants has been found to result in a life-time risk of major bleedings, ranging from 1.4 to 3% per year, in clinical trials [27,28] (which had already excluded patients at a high risk of bleeding). Moreover, the incidence of major hemorrhage for patients ≥80 years of age has been estimated at 13.1/100 person-years, and these patients are not usually enrolled in clinical trials [29].A significant proportion of patients with AF, i.e. 30-50%, do not receive anticoagulation due to contraindications or a perceived risk or fear of treatment-induced bleedings [30,31]. Moreover, coherence on anticoagulation with warfarin for secondary prevention after stroke may be decreased to only 45% after 2 years [32].The risk/benefit ratio of LAA closure versus warfarin for the prevention of stroke, systemic embolism and cardiovascular death in nonvalvular AF was studied in a meta-analysis that included 2,406 patients with 5,931 patient-years (PY) of follow-up from the PROTECT AF (WATCHMAN Left Atrial Appendage Closure Technology for Embolic Protection in Patients with AF) trial and the PREVAIL (WATCHMAN LAA Closure Device in Patients with Atrial Fibrillation versus Long-Term Warfarin Therapy) trial and their respective registries [33].In patients with nonvalvular AF at an increased risk for stroke or bleeding who are candidates for chronic anticoagulation, LAA closure resulted in improved rates of hemorrhagic stroke (0.15 vs. 0.96 events/100 PY), cardiovascular/unexplained death (1.1 vs. 2.3 events/100 PY) and nonprocedural bleeding (6.0 vs. 11.3%) compared to warfarin [33].In the pooled analysis of the 2 randomized clinical trials that compared the WATCHMAN device, LAA closure significantly reduced bleeding beyond the procedural period, particularly when adjunctive anticoagulation was discontinued [33].The favorable effect of LAA closure on long-term bleeding should be considered when selecting a stroke prevention strategy for patients with nonvalvular AF [34].A 2-center registry (110 patients aged 71.3 ± 9.2 years; CHADS2 score 2.8 ± 1.2, CHA2DS2-VASc score 4.5 ± 1.6 and HAS-BLED score 3.8 ± 1.1) showed that stroke and major bleeding rates with LAA closure were significantly lower than the PROTECT AF results. Thus, LAA closure in nonvulvular AF in a real-world setting may result in lower stroke and major bleeding rates than what is reported in clinical trials [35].LAA closure achieved cost parity versus dabigatran (110 mg) at 4.9 years and versus warfarin at 8.4 years. At 10 years, LAA closure was cost-saving against all therapies [35].LAA closure in both settings achieves cost parity in a relatively short period of time and may offer substantial savings compared with current therapies. Savings are most pronounced among higher-risk patients and those unsuitable for anticoagulation.Moreover, as to cost effectiveness, there are additional supportive indications from the PROTECT AF trial [36].The initially used percutaneous LAA transcatheter occlusion (PLAATO) device was rigid and required 20-50% oversizing when compared with the LAA orifice, in order to achieve a stable position. In contrast, more recent devices (WATCHMAN and ACP) are more flexible and need approximately only 10-20% oversizing. The amount of oversizing has implications since LAA has an oval orifice [37,38]. The flatter shape of the more recent devices facilitates occlusion of LAAs that have a short proximal portion and an early separation into lobes.Due to the fact that, in 80% of cases, the LAA is multilobulated [10] with the number of lobes varying from 1 to 4, apart from substantial differences in length and orifice size, a flatter LAA occlusion device is appropriate. LAA 3D printing offers accurate modeling to select the size, geometry and type of device.Pericardial EffusionThe most frequent primary safety event in the intervention group in PROTECT AF was serious pericardial effusion (defined as the need for either percutaneous or surgical drainage), which occurred in 22 (4.8%) patients [39]. Pericardial effusion rates decline with increasing operator experience. The proper anatomical delineation and sizing of the device provided by 3D printing evaluation could contribute to further elimination of the hemodynamic severity of pericardial effusions, likely reducing the mechanical stretch applied to the tissues.Device EmbolizationDevice embolization occurred in 3/408 patients in PROTECT AF. The stability of the LAA device may also be improved by LAA 3D printing evaluation.A recent analysis of the nonrandomized Continued Access Protocol (CAP) Registry which included 460 subsequent patients after the PROTECT AF study, documented a significant improvement in the safety of the WATCHMAN LAA closure, as a result of the increased experience of the operators as well as technical improvements to the device [39]. In this group, serious periprocedural pericardial effusion was decreased, being observed in only 10 patients (2.2% overall) and with no procedure-related strokes reported. Iatrogenic Atrial Septal DefectsPROTECT AF has shown that iatrogenic atrial septal defects, frequently observed after the transseptal procedure with a large-diameter transseptal 12-Fr sheath, have a high spontaneous closure rate and are not associated with an increased rate of stroke or systemic embolization during long-term follow-up [40].LAA 3D printing visualization might eliminate septal trauma by providing detailed anatomical information for the proper orientation of septal puncture.In PROTECT AF, after the periprocedural time frame, ischemic stroke occurred in 9/408 subjects in the intervention group (1.3 events/100 PY) compared with 6 in the control group (1.6 events/100 PY) [41].However, in the meta-analysis that included 2,406 patients with 5,931 PY of follow-up from the PROTECT AF and PREVAIL trials and their respective registries, there were more ischemic strokes in the device group (1.6 vs. 0.9 and 0.2 vs. 1.0 events/100 PY) [33].In a small percentage of patients, thrombus may be localized on the device. In the majority of these patients, the detected thrombus disappears after short-term anticoagulation [39]. According to the analysis, a device-associated thrombus was found in 4.2% of the successfully implanted patients; of these, 17/20 (85%) were either asymptomatic or endothelialized with anticoagulation. This created an estimate for the device-related thrombus-associated stroke rate of 0.3%/100 PY.LAA device-related thrombus (DRT) is poorly defined and understood. Stroke, peripheral embolism or cardiac/unexplained death occurred in subjects with DRT at a rate of 3.4/100 PY of follow-up, intermediate in frequency between overall event rates previously reported for the device and warfarin arms in PROTECT AF [42]. DRT was identified on at least 1 TEE in 27 PROTECT AF patients, indicating a DRT incidence of 5.7%.However, a source of embolism other than the LAA cannot be excluded. It should be noted that in the SPAF III (Stroke Prevention in Atrial Fibrillation III) trial, the presence of complex aortic plaques on TEE was independently associated with an increased thromboembolic risk, supporting evidence that the cause of stroke is potentially multifactorial in elderly patients with AF. Thus, in this hypothetical scenario, LAA closure might be unlikely to prevent all ischemic strokes in these specific substrates [43].The issue of incomplete occlusion of the LAA has been reported even when the occlusion has been performed surgically. With any surgical technique, the incomplete occlusion ranges from 10 to 72% [44,45]. The effect of incomplete surgical occlusion of the LAA on the risk of thromboembolism is unclear and has been derived from small studies with different closure techniques (e.g. Modified-Maze procedure, coronary bypass or valvular surgery) and is probably related to the inherent thromboembolic risk of surgical patients [44,45]. There are data supporting that there is no difference of stroke rate in patients with successful closure or incomplete closure [46].During percutaneous LAA closure for stroke prophylaxis, the geometric variability of the LAA ostium may result in incomplete sealing of the LAA. Incomplete sealing could enhance thrombus formation and embolization of thrombi around the device into the circulation.The feasibility for the proper application of current devices is extremely high, based on MDCT measurements of the LAA. Mean TEE-like MDCT LAA diameter estimation may result in feasibilities for proper application of 93.9, 97 and 99.0% for the WATCHMAN, the ACP and either 1 of these 2 devices, respectively [47]. However, in practice, it appears difficult to achieve such a success target based on contemporary imaging facilities.In one study, cardiac CT follow-up detected any persistent LAA contrast filling in 62% of patients, but leak-sizes were small (1.5 ± 1.4 mm) [48]. Peri-device leaks were almost exclusively localized at the posterior portion of the LAA orifice (>90%). TEE follow-up revealed a peri-device flow in 36% of the patients (jet-sizes: ≤4 mm).In another study, TEE follow-up revealed that 32.0% of implanted patients had at least some degree of peri-device flow at 12 months [49]. TEE revealed either minimal or no peri-device flow (jet ≤5 mm width). The impact of peri-device flow severity, defined as minor, moderate or major (<1 mm, 1-3 mm and >3 mm, respectively) on the composite primary efficacy end point (stroke, systemic embolism and cardiovascular death) was estimated. Compared to patients with no peri-device flow, the hazard ratios were 0.85, 0.83 and 0.48 for minor, moderate and major peri-device flow, respectively.From these data, it seems that residual peri-device flow into the LAA after percutaneous closure with the WATCHMAN device is common and is not associated with an increased risk of thromboembolism. However, this finding should be interpreted with caution as the low event rate decreases the confidence of this conclusion.LAA 3D printing visualization will contribute to eliminate incomplete sealing and any destructive thrombotic contribution of stagnant flow around misaligned LAA and device surfaces.Since LAA closure seems to be a rapidly evolving practice, LAA modeling by 3D printing may be a proper means to facilitate the process and increase the accuracy of the intervention, thus becoming an example of applied personalized medicine in structural interventions.