Literature Sharing | Application of 3D Printing Technology in the Diagnosis and Treatment of Congenital Heart Disease
【Foreword】
Focusing on innovations in the diagnosis and treatment of congenital heart disease (CHD), 3D printing technology has emerged as a new force for optimizing surgical repair and intervention strategies, thanks to its advantage in accurately visualizing complex anatomical structures. While domestic research in this field continues to advance, the international community is also increasingly recognizing its clinical translation value. This paper systematically reviews and analyzes the applications of 3D printing technology in CHD.
Source: Battieux C, Haidar MA, Bonnet D. 3D-Printed Models for Surgical Planning in Complex Congenital Heart Diseases: A Systematic Review. Front Pediatr, 2019, 7:23.
【Abstract】
Background: 3D printing is an emerging technology supporting the field of congenital heart disease (CHD). The primary objective of 3D-printed models is to better analyze complex anatomical structures, thereby optimizing the formulation of surgical repair or intervention strategies.
Methods : We conducted a systematic review to assess the accuracy and reliability of CHD modeling and 3D printing, as well as proof-of-concept evidence for the benefits of 3D printing in planning interventions.
Results : Correlative studies demonstrated good agreement between this technology and anatomical measurements. Therefore, the technique can be considered reliable, limited by the subjectivity of the operator in modeling the defects. Case series studies showed that 3D printing offers advantages in depicting vascular anatomy and guiding surgical approaches. For complex cardiac anatomy, 3D-printed models have been proven useful in developing cardiac repair strategies. However, evidence-based data demonstrating the efficacy of 3D-printed models in improving outcomes following surgical or interventional treatment for CHD remains lacking, due to the difficulty in designing prospective studies with comprehensive and clinically well-defined endpoints.
Conclusions : 3D printing technology can be used to enhance understanding of complex CHD anatomy and guide surgical strategies.
【Introduction】
The advantages of 3D printing technology in improving the accuracy of anatomical diagnosis in congenital heart disease (CHD) have attracted increasing attention. Various aspects of this technology help address the high variability and complexity of CHD anatomy. However, the value of 3D printing in optimizing surgical or catheter-based interventions remains to be explored to date. At present, it has become feasible to establish 3D printing laboratories in tertiary surgical centers, enabling full-process local management from processing Digital Imaging and Communications in Medicine (DICOM) files to the final production of 3D-printed models.
【Methods】
This study focuses on the application value of 3D printing technology in surgical planning for complex congenital heart disease. Through systematic retrieval and analysis of relevant literature, it verifies the concept of the technology’s accuracy, reliability, and clinical application potential.
The databases searched include PubMed and Google Scholar. The search term combination rule required the inclusion of at least three of the following: “3D printing”, “3D-printed models”, “complex congenital heart disease”, and “surgery”. Terms such as “surgical planning” and “measurement correlation” could be selectively added according to research needs.
Literature screening was independently conducted by two congenital heart disease specialists. Studies involving only medical understanding of CHD or those using 3D-printed models solely for patient communication were strictly excluded. All articles describing the formulation of CHD surgical strategies or anatomical measurement correlation were included for in-depth analysis.
【Results】
1. Accuracy and Reliability of 3D-Printed Models in Congenital Heart Disease
The primary step in validating 3D printing technology is to confirm the reproducibility of 3D-printed models relative to the original anatomical structures through measurement accuracy studies. Analyzing the complete manufacturing workflow of 3D-printed models is critical for understanding potential measurement deviations during segmentation and printing, while the potential risks of discrepancies between the final 3D-printed models and the original anatomy must be carefully evaluated.
2. Pre-segmentation and Segmentation Proces
The first step in creating a 3D-printed heart model is the segmentation of raw data from various imaging modalities, including cardiothoracic computed tomography (CT), magnetic resonance imaging (MRI), and transesophageal echocardiography (TEE). Segmentation is the core process that converts DICOM images from a series of slices into a 3D scene, requiring individual isolation and 3D visualization of each structure (e.g., blood pool, myocardium, great arteries, coronary arteries, etc.).
In the pre-segmentation phase, algorithms process pixels to isolate selected structures; common methods include thresholding, edge attraction, clustering, and classification. The subsequent segmentation process fills the isolated structures into 3D space, transforming pixel slices into 3D models with accurate X, Y, and Z coordinates. Currently, segmentation is mostly performed in a semi-automatic mode, allowing modification of pre-segmentation parameters. After the operator defines the structures to be segmented, the algorithm completes the subsequent workflow.
3. Different Segmentation Software and Printing Method
A variety of segmentation software is currently used in clinical practice, and most require manual intervention to generate the final 3D -printed model, which inevitably introduces uncertainty. All coordinate information of the 3D-printed model is presented in a mesh format, and the most commonly used file format is Standard Tessellation Language (STL), which is also suitable for the final printing step.
The final stage of printing may also lead to discrepancies between the original data and the printed product, which are closely related to the printer type and printing technology. Common printing methods include Fused Deposition Modeling (FDM), Stereolithography, Selective Laser Sintering (SLS), PolyJet, and others. The reliability of the final 3D-printed model depends on a multi-step workflow, and cumulative errors or approximations may occur at almost every stage.
4. Selection of Segmentation Software The choice of segmentation software depends on user needs and familiarity with 3D segmentation and modeling. Currently, 3D modelers are mainly divided into two categories: The first category consists of general users, who tend to select predefined, auto-segmentation "turnkey" products. Such software integrates full-process functions and can generate the final 3D-printed model on a single platform (e.g., Mimics by Materialize). It offers the advantages of convenient operation and high efficiency, but has limitations such as high licensing fees (over 10,000 US dollars per year) and a lack of user autonomy in modeling, post-processing, and pre-printing preparation. The second category consists of self-taught users, who mainly use open-source software (e.g., Itk-snap, 3D Slicer). They control each step of segmentation by manually adjusting DICOM contrast or thresholds, and can flexibly manage the propagation and post-processing of segmentation results. Such software is free and regularly updated, but has a certain learning curve and may introduce subjective deviations during the modeling process.
5. Case Series of Surgical Planning Using 3D-Printed Models: Current Research ProgressThree series of studies have attempted to demonstrate the advantages of 3D-printed models in surgical planning for congenital heart disease (CHD): One prospective single-center study analyzed 35 cases of tetralogy of Fallot by comparing 3D-printed models with intraoperative findings. It confirmed that 3D-printed models can reliably predict intraoperative conditions. However, the study focused on defect types with high consensus in surgical decision-making, resulting in limited applicability of the findings. A multicenter prospective study used a case-crossover design. For complex CHD including double-outlet right ventricle, complex transposition of the great arteries, criss-cross heart, and single ventricle, a multidisciplinary team evaluated surgical indications based on conventional imaging and 3D-printed models separately, and the results were finally compared with surgical outcomes. The study found that 3D-printed models helped optimize surgical plans in 50% of cases, but did not identify the key anatomical structures that influenced decision-making. Moreover, the inclusion of multiple anatomical types made it difficult to predict application prospects. A prospective comparative study evaluated the effect of 3D-printed models in 25 patients with double-outlet right ventricle. The results showed that the group using 3D-printed models had significantly shorter ventilator support time and ICU length of stay, suggesting that 3D-printed models may optimize surgical strategies or reduce procedural time. However, the study was limited in the description of patient selection and random grouping.
【Conclusions】
3D modeling and 3D printing technologies demonstrate favorable reliability and are suitable for anatomical analysis and surgical planning in congenital heart disease (CHD). However, the technology still has limitations and potential deviations to date, which are mainly related to the subjectivity involved in the CHD modeling process. Currently, it can only serve as an auxiliary tool for clinical decision-making in complex congenital heart disease. In the future, efforts should focus on optimizing the acquisition and operational workflow of 3D-printed models in daily clinical practice and lowering the barrier to utilization. The ultimate goal is to develop 3D-printed models into a diagnostic summary carrier for individual patients and a reliable tool for treatment strategy formulation. Achieving this goal requires the accumulation of evidence-based medicine data through more studies with well-defined clinically meaningful endpoints.
