ADVERTISEMENT
3D Printing for Mesenteric Artery Endovascular Interventions: Feasibility and Utility for Preprocedural Planning and Angiographic Correlation
Abstract
Background. Three-dimensional (3D) printing of mesenteric artery (MA) anatomy preprocedurally for endovascular interventions can allow strategic preprocedure planning and improve procedure-related clinical outcomes. Methods. Three patients with computed tomography angiography (CTA) of the abdomen and pelvis who subsequently underwent MA interventions were 3D printed retrospectively, and 2 patients with symptoms and severe MA stenosis on CTA, who had not undergone intervention, were 3D printed for procedure-related planning and anatomy-specific implications. The 3D-printed models (3D-PMs) were painted with acrylic paint to highlight anatomy. Reference vessel size, lesion length (LL), and renal artery (RA) to MA distance were determined using a digital millimeter caliper. Results. Each of the 5 patients with variable anatomy, including an MA chronic total occlusion (CTO), were successfully 3D printed. A digital caliper allowed determination of vessel size, LL, and RA to MA distance, which were then compared with intraprocedural MA angiograms and intravascular imaging when available. Further complex anatomies, such as intraprocedural navigation in the setting of prior abdominal aortic endograft and CTO assessment with relevance to cap morphology, small branch arteries, and collateral flow, were also successfully 3D printed. Conclusion. Preprocedural 3D printing of MA anatomy for interventions can theoretically lead to decreases in contrast use, radiation dose, and fluoroscopic and procedural times, as well as enhance comprehension of complex patient-specific anatomy.
J INVASIVE CARDIOL 2022;34(7):E510-E518. Epub 2022 May 6.
Key words: 3D printing, chronic total occlusion, mesenteric artery, procedure planning
Mesenteric ischemia can be classified as either acute or chronic, and while the former is of sudden onset requiring emergent intervention, diagnosis is often delayed in the latter, with patients often waiting months to years before seeking medical attention. This delay in diagnosis is related to robust arterial supply and collateral filling with development of symptoms when 2 of the 3 mesenteric arteries (MAs) are affected and include celiac artery (CA), superior mesenteric artery (SMA), or inferior mesenteric artery (IMA).1 A peak systolic velocity of >275 cm/s on duplex ultrasound (DUS) is highly specific to a stenosis of >75%; however, accurate diagnosis is limited by body habitus, bowel gas obscuring image acquisition, and operator variability and expertise. Therefore, computed tomography angiography (CTA) or magnetic resonance angiography (MRA) have become the diagnostic noninvasive imaging modalities of choice.2 Since MRA is limited by time, patient tolerability, and exclusion in the setting of incompatible metal foreign body, CTA is predominantly used. A prospective, comparative study between DUS, CTA, and MRA found mean image quality superior with CTA and MRA compared with DUS, and CTA with the highest sensitivity, specificity, and best diagnostic accuracy for stenosis grading.3 Three-dimensional (3D) volumetric reconstructed CTA can enhance the comprehension of anatomy, as it allows for the appreciation of anatomy in a 3D format; however, it is still limited by being a virtual image. 3D printing of MA anatomy can overcome this virtual barrier, allowing augmented tactile, sensory, and perceptual feedback not provided by the aforementioned diagnostic modalities. In this article, we discuss the feasibility and utility of 3D printing of MA anatomy for endovascular interventions.
Methods
Three patients with CTA who also subsequently underwent MA (including CA or SMA) interventions were queried. Two patients with symptoms and CTA indicating severe MA stenosis but awaiting MA interventions were also 3D printed to investigate preprocedural planning in a prospective, real-life scenario going forward. Digital imaging and communications in medicine (DICOM) images from abdominal CTAs were downloaded into the 3D slicer software (Surgical Planning Laboratory), the relevant MA and renal artery (RA) anatomy was segmented, and a computer-aided image was created. The digital images were 3D printed using Ultimaker 3 software (Create Education Limited) and Ender 3 Pro printers (Creality 3D). The material used for models was polylactic filament using a fused-deposition modeling method, which allows heated filament extrusion in the form of the computer-aided image, solidifying at room temperature. The 3D-printed models (3D-PMs) were painted with acrylic paint, with the aorta in yellow, MA and RA in red, venous flow in blue, and the endograft (Case 3 only) in teal. Reference vessel size (RVS), lesion length (LL), and RA to MA distance were extracted for clinical preprocedural planning. Moreover, in chronic total occlusion (CTO) MA lesions, 3D-PM was used to assess the CTO cap and collateral flow to the distal vessel. Comparisons of the 3D-PMs were systematically made with MA angiograms and 3D-CTA, including intravascular ultrasound (IVUS) imaging when available.
Results
Patient clinical presentation, DUS, and CTA findings are presented in Table 1 and procedure-related characteristics are shown in Table 2.
Case 1: severe proximal SMA and ostial CA stenosis. The patient was a 67-year-old female with severe proximal SMA and ostial CA stenosis. The 3D-PM shows labeled MA anatomy, an orthogonal view, and close up of stenosis (Figures 1A-1C). 3D-PM of the SMA demonstrated an RVS of 6.3 mm and LL of 16.9 mm (Figures 1Da, 1Db). While the CA stenosis was not treated during the index procedure, CA-RVS of 5.9 mm and CA-LL of 13.7 mm were obtained (Figures 1Dc, 1Dd). Intraprocedural IVUS depicted CA-RVS and SMA-RVS of 6.0 mm each. The SMA stenosis was predilated with a 5.0 x 12-mm balloon and treated with a 6 x 22-mm balloon-expandable covered stent. Abdominal aortogram, selective SMA angiogram, intervention, and postintervention angiograms are shown in Figures 1E-1G.
Case 2: critical ostial CA stenosis and CTO of SMA. The patient was a 68-year-old male with severe ostial CA stenosis and CTO of the ostial SMA. Smaller branch arteries and collateral arteries were also 3D printed to evaluate sensitivity and feasibility. As seen on the 3D-PM, smaller branch arteries and collateral flow were successfully 3D printed and relevant anatomy was labeled, including an orthogonal view (Figures 2A, 2Ba). CA-RVS of 8.4 mm was obtained with a CA-LL of 17.8 mm (Figures 2Bb, 2Bc). Intraprocedurally, a 5 x 20-mm balloon was used for predilation and a 7 x 22-mm covered stent was deployed (Figures 2Eb, 2Ec). Figure 2Ca and Figure 2Cb show a close-up view of the ostial stump of the SMA-CTO with a tapered distal cap. SMA-CTO RVS of 7.0 mm and LL of 21.6 mm were obtained, and intraprocedurally, a 4 x 20-mm balloon was used for predilation with deployment of a 6 x 22-mm covered stent. Another potentially significant measurement that can be obtained on 3D-PM is RA to MA distance. This can be especially important in complex anatomic cases (difficult to engage ostial CTO cap) or if limited contrast use is warranted in cases of renal insufficiency. A right RA to CA ostium distance of 27.0 mm and right RA to ostial SMA-CTO cap distance of 20.4 mm was obtained (Figures 2Da, 2Db). After successful CA intervention (Figures 2Ea-2Ed), SMA-CTO intervention was attempted via retrograde wire escalation (Figure 2Ee) but was unsuccessful even after obtaining a second arterial access to allow dual-contrast angiography. A second attempt was made with antegrade wire escalation and subsequent subintimal wire entry (Figure 2Fa) and successful recanalization with re-entry device (Figure 2Fb). SMA angioplasty, stent placement, and final angiogram postintervention are shown (Figures 2Fc, 2Fd, 2G).
Case 3: severe ostial CA artery and CTO of the SMA with prior endograft. The patient was a 78-year-old female with prior endograft, severe ostial CA stenosis, and suspected thrombosis of a proximal SMA stent. The operator noted significant difficulty in finding the CA ostium adjacent to the endograft. 3D-PM with labeled anatomy demonstrates proximal CA stenosis and proximal blunt SMA-CTO cap with a tapered distal SMA-CTO cap (Figures 3A, 3B). CA-RVS of 6.8 mm and CA-LL of 9.2 mm were obtained on the 3D-PM (Figures 3Ca, 3Cb). Intraprocedurally, predilation was accomplished with a 5.0 x 12-mm balloon with placement of a 6- x 10-mm bare-metal stent. A left RA to CA ostium distance of 23.4 mm and left RA to proximal SMA-CTO cap distance of 14.5 mm were obtained on the 3D-PM (Figures 3Cc, 3Cd). Although not demonstrated, a proximal edge of the endograft to the CA or SMA ostium can also be obtained on the 3D-PM for intraprocedural navigation and easier identification of the MA ostium by catheter engagement. CA and MA angiogram, intervention, and postintervention angiograms are depicted (Figures 3D-3F).
Case 4: moderate CA and moderate to severe SMA ostial stenosis on CT scan. The patient was an 85-year-old symptomatic female with moderate CA and moderate to severe ostial SMA stenosis on CT scan. While this patient did not undergo MA intervention, a 3D-PM was printed to demonstrate real-life implications of 3D printing preprocedurally in anticipation of a future planned procedure. 3D-PM with labeled relevant anatomy is shown along with orthogonal views (Figures 4A-4C). The 3D-PM shows that the SMA ostial stenosis appears worse in severity than the CA stenosis. Right RA was also printed, as accessory arteries were noted on CT scan. CA-RVS of 6.9 mm and CA-LL of 10.8 mm (Figures 4Da-4Db) along with SMA-RVS of 6.9 mm and SMA-LL of 9 mm were obtained (Figures 4Dc-4Dd). Left RA to SMA distance of 14.6 mm and right renal superior accessory artery to CA distance of 32.6 mm were obtained (Figure 4E). A distal abdominal aortogram obtained at the time of the prior lower-extremity peripheral intervention along with 3D-CTA are shown for comparison (Figure 4F).
Case 5: proximal CA stenosis on CT scan. The patient was an 87-year-old symptomatic female with proximal CA stenosis on CT scan. 3D-PM with labeled anatomy and an orthogonal view are shown (Figures 5A, 5C). An acute 90° bend is noted after approximately 1 cm of the CA-LL, which is highlighted (Figure 5B). CA-RVS of 7.2 mm and CA-LL of 19.6 mm were obtained (Figures 5Da, 5Db). Additionally, left RA to CA ostium distance of 54.8 mm was determined (Figure 5E). 3D-CTA is shown for comparison (Figure 5F).
Discussion
Chronic mesenteric ischemia (CMI) is characterized as insidious onset with gradual progression, clinically presenting as postprandial abdominal pain lasting for 30 minutes with associated weight loss, food apprehension, nausea, vomiting, and/or diarrhea.4 Atherosclerosis involving the proximal CA, SMA, or IMA is the most common cause, while less-common etiologies include dissection, vasculitis, fibromuscular dysplasia, postradiation, and cocaine abuse.4 CMI is considered a disease of advanced age, typically presenting in patients older than 50 years with concomitant cardiovascular disease risk factors (ie, hypertension, hyperlipidemia, diabetes mellitus, renal disease, tobacco abuse, obesity, and/or sedentary lifestyle).5 Endovascular intervention (EVI) is the preferred revascularization strategy over open surgical repair because of lower associated morbidity and mortality. 3D printing of MA anatomy preprocedurally can be an important benefit in permitting enhanced understanding of complex anatomy and culprit lesion and appreciation of vessel tortuosity and ostium take-off. Furthermore, it allows preprocedure determination of access-site choice, catheter selection, and assessment of CTO cap (blunt vs tapered) with distal vessel filling via collaterals.
Evidence for EVI in CMI. As discussed, EVI is preferred over surgical repair for CMI as the patient population experiencing the disease process is often frail and unable to tolerate surgery with baseline comorbidities. Although MA-EVI is associated with lower morbidity, mortality, and shorter hospital stay, it has been associated with higher reintervention rates compared with surgery.6,7 In 1 consecutive series of 59 patients who underwent MA-EVI, 96% obtained procedural success at a mean follow-up of 38 months and 17% experienced recurrence of symptoms and underwent repeat intervention for restenosis.8 In another study comparing 3-year outcomes between surgery vs EVI, no differences in 30-day mortality, in-hospital complication rate, or 3-year survival rate were found between the 2 arms. However, freedom from recurrent symptoms was significantly lower within the surgical group compared with EVI (66% vs 27%; P<.02). This result can possibly be related to less complete revascularization in EVI, as the 1- and 2-vessel revascularization rates were 79% and 21% compared with 36% and 64% in the surgical group, respectively.9 One reason for the lower complete revascularization rate in the EVI group is the presence of CTOs, which are technique- and operator-experience dependent, and require expertise with the navigation of the hybrid algorithm to achieve a high procedural success rate.10 In a study by Sarac et al, 28% of 87 MA lesions treated for CMI were CTOs.11 As demonstrated in Case 2, CTO cap morphology as well as collateral and distal vessel filling appreciation can be significantly enhanced with 3D-PM of MA, improving operator confidence for procedural success. As percutaneous transluminal angioplasty (PTA) is associated with elastic recoil and vessel closure, adjunctive stenting is predominantly employed for MA-EVI. However, evidence in the literature gives preference to covered stents compared with bare-metal stents. In a nonrandomized study, covered stents had higher primary patency rates than bare-metal stents at 3 years (92 ± 6% vs 52 ± 5%; P<.01). Furthermore, freedom from restenosis (92 ± 6% vs 53 ± 4%; P<.01), recurrence of symptoms (92 ± 4% vs 50 ± 5%; P<.01), and reintervention rates (91 ± 6% vs 56 ± 5%; P<.01) at a mean follow-up of 29 ± 12 months also favored the covered stent arm. Additionally, in those requiring reinterventions (followed to 2 years), higher rates of freedom from restenosis, symptom recurrence, and reintervention were still seen within the covered stent group. Given this body of evidence, covered stents for revascularization were more commonly employed within our small, retrospective subset.12
Evidence for benefits of 3D printing for percutaneous interventions. The realm of structural heart disease has seen significant progress toward the incorporation of preprocedural 3D printing technology for both anatomic and physiologic assessment. Current applications of 3D printing for structural interventions, for example, include left atrial appendage anatomy for the selection of appropriate occluder device;13,14 aortic valve models to predict coronary height, paravalvular leak anatomic location, and assessment of flow dynamics, orifice area, and calcium leaflet deposits;15,16 mitral valve anatomic models to predict left ventricular outflow tract obstruction and optimal valve seating;17 and atrial septal defect models to predict adequate tissue rim for device deployment.18 Recently, the role of 3D printing for the percutaneous coronary intervention of complex coronary artery disease in bypassed patients has been evaluated, with preprocedural 3D-PMs allowing augmented understanding of difficult-to-find occluded bypass graft buttons; appropriate catheter selection for difficult-to-engage bypass graft origin; ostial CTO cap morphology; and distal vessel filling to aid with hybrid algorithm navigation for CTO revascularization—all of which can improve clinical outcomes, translating to decreased contrast volume use, radiation dose, and procedural and fluoroscopic times.19 Preprocedural 3D printing of aortic arch and carotid anatomy for endovascular carotid stent procedure allows for the appreciation of aortic arch type, leading to appropriate access site and catheter selection; improved understanding of common carotid to internal carotid artery angulation; landing zone assessment predicting success of embolic protection device deployment; and preprocedural selection of balloon, stent, and embolic filter size.20 Additionally, the benefits of 3D printing for RA interventions has been recently investigated. In this study, the authors discuss the advantages of 3D-printed RA models to enhance operator understanding of the RA ostium take-off for optimal catheter engagement; preprocedural balloon and stent size selection; and appreciation of diverse renal pathology including fibromuscular dysplasia, ostial atherosclerotic, and bilateral RA stenosis.21
Feasibility and utility of 3D printing of MA anatomy preprocedurally for EVI. 3D printing can be useful with preprocedural planning of MA-EVI. 3D printing of the MA allows for the selection of appropriate predilation and postdilation balloons and stent sizes by the measurement of the RVS, as illustrated in all 5 3D-PM cases. Additionally, RA to MA distance can be obtained preprocedurally, allowing the operator to comprehend the location of the MA ostium and leading to intraprocedural decreases in contrast volume, radiation dose, and procedure and fluoroscopic times. In patients with CTO, proximal cap anatomy (blunt vs tapered) as well as collateral and distal vessel filling anatomy can allow easier navigation of the hybrid algorithm and allow for a more efficient and successful procedural outcome.10 3D printing of collateral arteries and small branch arteries, such as the iliojejunal and ileocolic arteries, is also feasible, as demonstrated in Case 2. Assessment of MA ostium angulation and/or vessel tortuosity can also allow preprocedural access-site and catheter selection for optimal engagement and support. Finally, in individuals with complex anatomy, such as those with endografts as in Case 3, the orientation of the MA ostium relative to the endograft (endograft to MA distance), along with determination of the MA ostium take-off, can be better appreciated preprocedurally.
Conclusion
The ability of 3D printing to accurately and precisely replicate patient anatomy has shown considerable utility with preprocedural planning for MA-EVI. Specifically, preprocedural planning of difficult anatomical subsets, including CTO, angulated ostium, distance relative to RA or aortic endografts (where present) for efficient catheter engagement, as well as the selection of balloons and stents can all be accomplished. Theoretical advantages include improved procedure-related clinical outcomes regarding contrast volume, radiation dose, procedure time, and fluoroscopic time. Prospective analysis in a randomized study is warranted to prove this hypothesis.
Affiliations and Disclosures
From the 1Division of Cardiovascular and Endovascular Disease, Einstein Medical Center, Philadelphia, Pennsylvania; and 2Division of Cardiovascular Research, Einstein Medical Center, Philadelphia, Pennsylvania.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.
The authors report that patient consent was obtained for the images used herein.
Manuscript accepted June 28, 2021.
Address for correspondence: Jon C. George, MD, Einstein Medical Center, 1200 West Tabor Rd, Tabor Medical Building - Moss, 3rd Floor, Philadelphia, PA 19141. Email: jcgeorgemd@gmail.com
References
1. Walker TG. Mesenteric ischemia. Semin Intervent Radiol. 2009;26(3):175-183. doi:10.1055/s-0029-1225662
2. Moneta GL, Lee RW, Yeager RA, Taylor LM Jr, Porter JM. Mesenteric duplex scanning: a blinded prospective study. J Vasc Surg. 1993;17(1):79-84; discussion 85-86. doi:10.1067/mva.1993.41706
3. Schaefer PJ, Pfarr J, Trentmann J, et al. Comparison of noninvasive imaging modalities for stenosis grading in mesenteric arteries. Rofo. 2013;185(7):628-634. doi:10.1055/s-0033-1335212
4. Hohenwalter EJ. Chronic mesenteric ischemia: diagnosis and treatment. Semin Intervent Radiol. 2009;26(4):345-351. doi:10.1055/s-0029-1242198
5. Casserly IP, Sachar R, Yadav JS. Practical Peripheral Vascular Intervention. 2nd ed. Lippincott Williams & Wilkins; 2012.
6. Pecoraro F, Rancic Z, Lachat M, et al. Chronic mesenteric ischemia: critical review and guidelines for management. Ann Vasc Surg. 2013;27(1):113-122. doi: 10.1016/j.avsg.2012.05.012
7. Hogendoorn W, Hunink MG, Schlösser FJ, Moll FL, Muhs BE, Sumpio BE. A comparison of open and endovascular revascularization for chronic mesenteric ischemia in a clinical decision model. J Vasc Surg. 2014;60(3):715-725.e2. doi:10.1016/j.jvs.2014.03.009
8. Silva JA, White CJ, Collins TJ, et al. Endovascular therapy for chronic mesenteric ischemia. J Am Coll Cardiol. 2006;47(5):944-950. doi:10.1016/j.jacc.2005.10.056
9. Kougias P, Huynh TT, Lin PH. Clinical outcomes of mesenteric artery stenting versus surgical revascularization in chronic mesenteric ischemia. Int Angiol. 2009;28(2):132-137.
10. Rangan BV, Kotsia A, Christopoulos G, et al. The hybrid approach to intervention of chronic total occlusions. Curr Cardiol Rev. 2015;11(4):299-304. doi:10.2174/1573403X11666150909113026
11. Sarac TP, Altinel O, Kashyap V, et al. Endovascular treatment of stenotic and occluded visceral arteries for chronic mesenteric ischemia. J Vasc Surg. 2008;47(3):485-491. doi:10.1016/j.jvs.2007.11.046
12. Oderich GS, Erdoes LS, Lesar C, et al. Comparison of covered stents versus bare metal stents for treatment of chronic atherosclerotic mesenteric arterial disease. J Vasc Surg. 2013;58(5):1316-1323. doi:10.1016/j.jvs.2013.05.013
13. Liu P, Liu R, Zhang Y, Liu Y, Tang X, Cheng Y. The value of 3D printing models of left atrial appendage using real-time 3D transesophageal echocardiographic data in left atrial appendage occlusion: applications toward an era of truly personalized medicine. Cardiology. 2016;135(4):255-261. doi:10.1159/000447444
14. Goitein O, Fink N, Guetta V, et al. Printed MDCT 3D models for prediction of left atrial appendage (LAA) occluder device size: a feasibility study. EuroIntervention. 2017;13(9):e1076-e1079. doi:10.4244/EIJ-D-16-00921
15. Qian Z, Wang K, Liu S, et al. Quantitative prediction of paravalvular leak in transcatheter aortic valve replacement based on tissue-mimicking 3D printing. JACC Cardiovasc Imaging. 2017;10(7):719-731. doi:10.1016/j.jcmg.2017.04.005
16. Maragiannis D, Jackson MS, Igo SR, et al. Replicating patient-specific severe aortic valve stenosis with functional 3D modeling. Circ Cardiovasc Imaging. 2015;8(10):e003626. doi:10.1161/CIRCIMAGING.115.003626
17. El Sabbagh A, Eleid MF, Said S, et al. 3D printing for procedural simulation of transcatheter mitral valve replacement in patients with mitral annular calcification. J Am Coll Cardiol. 2018;92(7):E537-E549. Epub 2018 Jan 23. doi:10.1002/ccd.27488
18. Chaowu Y, Hua L, Xin S. Three-dimensional printing as an aid in transcatheter closure of secundum atrial septal defect with rim deficiency: in vitro trial occlusion based on a personalized heart model. Circulation. 2016;133(17):e608-e610. doi:10.1161/CIRCULATIONAHA.115.020735
19. Memon S, Friend E, Samuel SP, et al. Patient-specific coronary artery bypass graft 3D-printing: implications for procedural planning in complex percutaneous coronary interventions. J Invasive Cardiol. 2021;33(8):E592-E603.
20. Memon S, Friend E, Samuel SP, et al. 3D printing of carotid artery and aortic arch anatomy: implications for preprocedural planning and carotid stenting. J Invasive Cardiol. 2021;33(9):E723-E729.
21. Memon S, Friend E, Kalra S, Janzer S, George JC. 3D printing of renal arteries for endovascular interventions: feasibility, utility and correlation with renal arteriograms. J Invasive Cardiol. 2021;33(12):E986-E992.
Related Articles
- 3D Printing of Subclavian Artery: Utility for Preprocedural Planning and Correlation With Subclavian Artery Percutaneous Vascular Interventions
- 3D Printing of Renal Arteries for Endovascular Interventions: Feasibility, Utility, and Correlation With Renal Arteriograms
- 3D Printing of Carotid Artery and Aortic Arch Anatomy: Implications for Preprocedural Planning and Carotid Stenting
- Patient Specific Coronary Artery Bypass Graft 3D-Printing: Implications for Procedural Planning in Complex Percutaneous Coronary Interventions
