Development and Evaluation of 3D-Printed Aortic Phantoms for Multimodal Patient-Specific Therapy Planning

Objectives: Individualization is becoming more and more important in cardiovascular surgery. Patient-specific models can help for an adequate patient-prosthesis matching as well as to plan an interventional or surgical procedure. The aim of this study was to develop a process for production and multimodal in-vitro assessment of patient specific 3D-printed aortic models.

Methods: Anonymized CT datasets from patients were collected retrospectively. Based on the datasets, patient-specific volumetric models were created. Different flexible 3D-printing materials were compared in terms of mechanical properties. Native aortic samples served as reference group. Based on tensile testing results an appropriate material was chosen, the model's wall thickness was set and different aortic models were printed.

Models were scanned by X-ray, CT, MRI, and (Doppler) sonography to assess suitability for clinical imaging. Contrast media was used to enhance visibility in x-ray based imaging and to mimic angiographic settings. For simulation of the intraluminal blood flow, a mock circulation with pulsatile flow profile was established. Sonographic Doppler imaging and 4D-MRI were used for flow visualization.

Results: A total number of 12 hollow 3D models representing different aortic anatomies and pathologies were manufactured. Mechanical properties were comparable to physiological biomechanics. Phantoms were translucent and showed a high resolution of 30 µm. Evaluation of both, digital and physical 3D models showed high accordance to the CT data regarding anatomical accuracy. Printed models were permeable for X-ray as well as ultrasound and visualization was comparable to human vessels. Doppler sonography displayed flow profiles matching native data. 4D-MRI flow measurements of aortic models showed landmark flow characteristics (i.e., low flow areas, turbulences, pressure asymmetry, and wall shear stress). These were comparable in size and location to the patient’s flow measurements, especially for pathologic anatomies like aneurysms.

Conclusion: It was possible to produce patient-specific aortic phantoms with physiologic mechanics. Multimodal assessment, using common clinical imaging techniques as well as 4D-MRI was possible. This new process is ideal for in vitro prosthesis evaluation and can be used in patient-specific therapy planning.