Customized Bone Allografts by 3D-printing (3D-MALF 2)

Customized Bone Allografts by Virtual Surgical Planning and 3D-printing for Correcting Musculoskeletal Deformities in Children

Virtual surgical planning (VSP), the simulation of bone corrections in virtual reality ("Computer Aided Surgical Simulation": CASS) and 3D printing of customized implants and devices are achieving an increasingly central role in clinical practice and orthopaedic surgery. Those technologies and processes allow an allow incredibly versatile and accurate planning and reproduction of complex bone correction or joint replacement procedures. Recent and converging evidence document how the use of these technologies is able to significantly reduce surgical times, bleeding and intra-operative complications, and the use of intra-operative fluoroscopy. Due to the collaboration between the ward of Pediatric Orthopedics and Traumatology of the Rizzoli Orthopedic Institute and the Department of Industrial Engineering (DIN) of the University of Bologna it was possible to experiment, validate and introduce simulation, planning and personalization technologies of interventions of corrective surgery of Musculoskeletal Disorders (MSDs) of the limbs in childhood and developmental age into clinical practice. (3D-MALF - CE AVEC: 356/2018/Sper/IOR). Currently, extremely complex bone correction interventions are often planned and performed through Computer Aided Design (CAD) and 3D printing of models and custom sterilizable cutting guides (Patient-Specific Instrument, PSI). In pediatric orthopedic surgery is often necessary to use homologous massive bone grafts customized on the patient's anatomy, which can be employed in the replacement of neoplastic lesions, in the axial correction of deformities or even in the extemporaneous lengthening of bone segments. The Musculoskeletal Tissue Bank (BTM) regularly provides bone grafts processed in a Class A controlled contamination environment according to GMP (Clean Room), guaranteeing quality and microbiological safety. The current realization standard of bone grafts on specific request is a freehand realization. The BTM technicians model the grafts, based on the indications received (length, width, height, indications on geometry), using standard surgical instruments (osteotomes, oscillating saws, etc.). The present clinical trial aims to validate the feasibility, accuracy and effectiveness of an innovative process for producing customized bone allografts to correct bone deformities in children. the customization process will be conducted by using computer-aided surgical simulation and 3D printing.

The study is a double-blind randomized clinical trial and point to achieve a comparison of geometry and use in the operating room between manually modelled bone grafts and bone grafts modelled using CAD techniques and customized cutting guides realized with 3D printing. In addition, two other objectives will be pursued: (1) to conduct a preliminary analysis of the management and production costs of customized bone grafts and of the generation process using VSP and 3D printing of models and surgical interventions aids such as sterilizable PSI and GSI; (2) to treat pediatric patients suffering from congenital malformations that require complex multiplanar, multifocal and/or multisegmental bone correction surgery planning the intervention using VSP and implanting bone grafts produced by BTM with the aid of sterilizable PSI made by 3D printing. The study involves the creation of a clinical-surgical path, divided into various consecutive steps implemented in part by PED-IOR, by BTM-IOR, by DIN-UNIBO and by Diagnostic and Interventional Radiology (RAD-IOR). The process begins with the selection of the patient for the study by PED-IOR. After a clinical evaluation of the patient and the definition of the underlying pathology and deformity to be treated, if the patient complies with the inclusion criteria, a hypothesis of treatment and a choice of surgical technique is identified. Then, the path is illustrated to the patient and the family, also for the purpose of acquiring a written informed consent. Once the patient is enrolled, the path includes the radiological study of the patient by RAD-IOR. As normal clinical practice requires, the bone segments will first be examined by panoramic x-rays with any comparatives of the contralateral limb, then the pre-operative CT scan at very low radiation dose will be performed. The diagnostic imaging data of the treated patients and of the bone segments that have to be processed from the bank, will be kept on a company network computer and pseudo-anonymized according to current company procedures in line with the recent law provisions on the protection of sensitive data (GDPR 2016/679). In particular, imaging data are renamed with a unique PatientID for each patient (the file that associates the patient's name with the PatientID will be password protected and kept on the company network). The pseudo-anonymized radiographic images will be segmented by PED-IOR on a computer on the hospital network, obtaining a file in "Standard Triangulation Language" (STL) named with the PatientID. The segmentation of the pre-operative CT will be obtained using a workflow validated by us for clinical and surgical use: the DICOM file is processed and converted using open-source software (3D-Slicer 4.11 or similar software with free license) into an STL file, which format file is completely anonymous and allows the three-dimensional display of the data. The file is now ready to be sent to DIN-UNIBO to realize surgical simulation, three-dimensional planning, 3D printing of models and PSI, according to the surgeon instructions. Three-dimensional surgical planning is performed on the STL file through a continuous comparison between the DIN-UNIBO engineer and the IRCCS-IOR surgeon. The surgeon hypothesizes and suggests the possible corrections to be made, as well as the design of PSI, that will be simulated on the computer using open source or with academic licenses software of design and three-dimensional modelling, such as Blender 2.8 or similar. The engineers will provide the "rendering" of the planned correction and PSI, which will form the basis for the preparation of the models, grafts and PSI/GSI if approved by the surgeon. Once the planning has been defined and models and PSI have been designed, the DIN-UNIBO engineers will proceed with the drafting of specific reports with the details of the intervention. They will also deal with the 3D printing of models and PSI, using the technology already validated and currently used in our unit for the execution of complex osteotomies. Specifically, models are converted into G-code printing code (Ultimaker Cura v 4.13.1, or similar) and then 3D-printed in High Temperature Polylactic Acid (HTPLA) using a 3D printer appropriately set. The entire process has already been validated by us and allows the production of models and cutting guides that can be sterilized with the current autoclave models present in the Institute at 134 ° C for 50 minutes. The geometry of the graft will be defined according to the surgical planning. The bone tissues, from which the grafts will be obtained, are taken from the cadaver donor, treated and stored in compliance with the guidelines "Minimum organizational, structural and technological requirements of the Institutes of tissues for quality and safety in donation, procurement, control, processing, conservation, storage and distribution of human tissues and cells "of the National Transplant Center. The musculoskeletal tissues distributed by BTM, regardless of the processing performed, are always included within normal clinical practice, as they are legally classified as transplants and not as medical devices. In this phase, the production method of the homologous bone graft will also be assigned, on the basis of the randomization criteria. After the virtual planning of the surgery, the main measures of the designed graft are extrapolated. Specifically, two groups will be defined: Standard group and GSI group. For the Standard Group, DIN-UNIBO will extrapolate from the graft design the main measurements (length, width, height, shape) and will send the two-dimensional drawing of the graft to BTM for the manual processing. The operators will perform the processing according to the routine procedures, obtaining the required graft on the basis of the information received. For the GSI group, the bone bank provides a series of CTs of bone segments from those available for processing. These bone segments are reconstructed and compared with the planning to find the best three-dimensional match between the planned bone graft and the available segments, possibly adapting the VSP schedule based on the graft selected. Subsequently, DIN-UNIBO will design models and GSI cutting guides and will 3D print them, and then will send them to BTM-IOR together with the indication of the segment to be processed. BTM-IOR will execute the sterilization of the GSI that will be used in the clean-room processing of the graft. The segments will be treated with an oscillating saw, using the customized models and cutting guides (GSI). The grafts obtained, regardless of the group they belong to, will be sterile packed according to the standard procedures BTM-IOR, including only an indication on the direction of the implant, in order not to give any indication to the surgeon about the preparation method. The surgery will be performed according to the pre-operative planning established, using the graft and the PSI produced for the specific case. To obtain a size and shape comparison between the planned graft and the graft made in the clean room, a three-dimensional reconstruction of the latter must be available on a computer. Since BTM-IOR must guarantee the sterility of the graft and that 3D scanning technologies suitable for use in clean rooms are not currently available in IOR, CT scanning of the packaged grafts will be realized, as the packaging is almost radio-transparent. The CT scans will be performed inserting the grafts in the bank tissue scan protocol and therefore no additional resources are required to perform this procedure by IOR. CT examinations of the grafts will be executed with GE Revolution 64 Slice Dual Energy equipment and with a layer thickness of less than 1 mm to have data available to allow reliable measurements and three-dimensional reconstructions. CT images of the graft will be used for 3D reconstruction, which will be employed to extract the main measurements through CAD software. These measurements are the most clinically significant, and will be compared with the same measurements of the planned graft. Moreover, having the geometry of the implanted graft available also makes it possible to perform a volumetric difference analysis using point cloud comparison software, such as Meshlab or CloudCompare. As an additional check of the correct production of the graft according to the schedule, the main dimensions of the graft will still be measured in the clean-room with a sterile gauge, immediately after processing and before packaging. The study duration will be four years from its approval. The follow up is fixed at one year. During the first three years of the study, the enrollment and execution of the interventions will take place, while the last year is intended for the conclusion of the follow-up of the last patients treated. The collected data will be analyzed using SPSS v. 22 A preliminary analysis and a descriptive report will be drawn to express: the geometric differences between planned and realized graft; demographic and clinical variables of patients; deformity type, location and degree as measured on panoramic radiographs and CT, comparing it with the counter-lateral limb (or with general population physiological parameters, in case of bilateral deformity); degree and characteristics of the planned correction using VSP; methods, timing, costs and problems related to the preparation of the VSP, selection of the graft to be processed, processing of the customized bone graft, printing of the instrumentation (such as cutting guides), operating times, blood losses, any intra-operative and peri-operative complications, problems related to cutting guides and bone grafting; correction obtained at post-operative control; clinical-functional outcome at one year of follow-up (the accuracy of the post-operative correction obtained comparing it to the planning through reliability analysis). Frequencies, means, medians, interquartile ranges and standard deviations will be used to describe the study variables. Differences between groups will be evaluated using the chi-square test for categorical variables and the t-Student test or Mann-Whitney U test for independent samples for continuous variables. No data collection and procedure/analysis will be conducted before signing the consent. Sensitive data will be processed in accordance with current legislation by the head of the study. Furthermore, participants can withdraw their consent to participate at any time, without any consequences. The protection of the personal information provided by the subjects will take place according to the current legislation on the protection of personal data. In line with international data protection legislation, the following measures will be taken: the paper material will be stored in dedicated lockers which are locked and not accessible to unauthorized persons; at the coordinating center IRCCS Istituto Ortopedico Rizzoli all data will be stored in digital format at the SC Ortopedia e Traumatologia Pediatrica with access allowed only with a password. All patients included will be identified with a numerical code to pseudo-anonymize sensitive. The data will be kept by the investigator for the time necessary for scientific production. In order to guarantee the confidentiality of clinical trial data as required by applicable national and European legislation, the data will be accessible only to the study promoter and his designees, for monitoring/auditing procedures, to the investigator and collaborators, to the Ethics Committee of the center where the research is conducted and to the relevant health authorities. The investigator and the Institute will allow access to the data and source documentation for monitoring, auditing, reviewing by the Ethics Committee and inspections by the Health Authority, but retaining the confidentiality of personal data in accordance with current legislation.

Bone allograft, 3D printing, Computer aided surgical simulation, Virtual surgical simulation, Orthopedics, Pediatrics, Additive Manufacturing, Cutting guides, 3D models

The study in defined as a randomized clinical trial with two groups and will be double-blind. The randomization will be based on the realization method of the grafts.

The main measures of the designed graft are extrapolated from VSP. The dimensioned drawings of the graft are delivered to the bone bank that prepares the graft according to standard procedures.

The main measures of the designed graft are extrapolated from VSP. The bone bank provides a series of CTs of bone segments from those available for processing. These bone segments are reconstructed and compared with the planning to find the best match. Once defined which bone segment will be used, if necessary, the planning is adapted to it.

Planning of multiplanar and/or multifocal osteotomies in children by using VSP, PSIs and customized structural bone allograft

Preoperative VSP is performed starting from 3D digital models obtained from DICOM data of CT-scan with a segmentation process. If applicable, a model of the contralateral limb is obtained and used as reference for the correction. The surgeon and the engineer together define the surgical strategy, establishing the correction, the design of Patient-Specific Instruments (PSIs) and the hardware to implant. Then, through VSP, an ideal model of the bone graft is designed.

The size dimensions obtained through VSP are used by the tissue bank to manually cut a structural bone graft according to the standard protocols.

The best donor bone is identified by matching the 3D models provided by bone bank with the planned graft, considering not only the size but also the most suitable configuration of cortical and medullary bone. Then GSIs are designed and 3D printed to perform the exact cut on the donor bone according to the experimental protocol.

The acute surgical correction is performed according to VSP using PSI and the customized bone graft. Preoperative variables (age, sex, disease, deformity), intraoperative variables (operation time, intraoperative complications, use of fluoroscopy, intraoperative bleeding) are registered.

The main dimensions of the obtained graft (height, lenght and depth), which are also the most clinically significant, will be measured in millimiters in the clean-room. The analysis of the dimensions obtained will be performed reconstructing the CT images of the graft. These dimensions will be compared with those planned by VSP, in order to examine the accuracy, reliability, and repeatability of the graft processing through GSI compared to standard processing.

The volume of the obtained graft will be measured in millimiters square in the clean-room. The analysis of the volumes obtained will be performed reconstructing the CT images of the graft. These values will be compared with those planned by VSP, in order to examine the accuracy, reliability, and repeatability of the graft processing through GSI compared to standard processing.

The cost analysis (calculated in Euros) will be performed according to a management model that includes pre-clinical assessments and imaging necessary for planning, production costs of custom bone graft, 3D printing equipment (e.g. cutting guides), the planning and related report and the costs related to the surgery. The resulting cost will be compared with that of traditional interventions considering in the evaluation the cost-effectiveness of the intervention also in terms of the correction obtained and any need for further procedures.

The maintenance of the geometrical-structural characteristics (height, lenght and width calculated in millimeters) of the post-intervention PSI will be checked. Once cleaned and treated according to the internal protocol (PG 19 DS. Rev 04 of 31/10/17 relating to the management of orthopaedic prosthesis explants), PSI will be further analysed and compared with the pre-operative geometries also by means of 3D scanning to evaluate the maintenance of the original design.

Intraoperative and post-operative blood loss (during the first 48 hours) will be estimated by using the López-Picado's formula

The number of possible intra and post-operative complications, which comprehend also the suitability of PSI and bone graft compared to the planned intervention, will be identified.

The obtained degree of correction, will be measured on standard anteroposterior and lateral radiographs of the operated bone segment.

To evaluate the clinical-functional outcome, the Paediatric Outcome Data Collection Instrument (PODCI) questionnaire will be administered pre-operatively and after one year of follow-up. This tool, recently validated in Italian, is a specific multidimensional questionnaire for paediatric patients with MSDs.

the number of additional surgeries for treating incomplete correction overcorrection or recurrent deformities will be calculated in both groups

Inclusion Criteria: From 2 to 18 years old Diagnosis of limbs musculoskeletal diseases Need for uni- or poly-axial correction through one or more osteotomies Need for massive bone graft to stabilize the correction Consent to the processing of data Exclusion Criteria: Patients who refuse the VSP study Patients who do not undergo in-depth radiological examinations or patients with insufficient radiological documentation Patients who undergo different interventions for the correction of musculoskeletal disease Patients who do not need a massive bone graft Pregnant women

We are planning to share preoperative IPD (age, sex, disease and deformity), intraoperative IPD (operation time, intraoperative complications, use of fluoroscopy, intraoperative bleeding) and outcomes.

The data will be kept by the principal investigator for the time necessary for scientific production.

IPD available on request due to restrictions. The IPD presented in this protocol could be available on request from the principal investigator. The data are not publicly available due to national privacy regulations.

Raza M, Murphy D, Gelfer Y. The effect of three-dimensional (3D) printing on quantitative and qualitative outcomes in paediatric orthopaedic osteotomies: a systematic review. EFORT Open Rev. 2021 Feb 1;6(2):130-138. doi: 10.1302/2058-5241.6.200092. eCollection 2021 Feb.

Grassi FR, Grassi R, Vivarelli L, Dallari D, Govoni M, Nardi GM, Kalemaj Z, Ballini A. Design Techniques to Optimize the Scaffold Performance: Freeze-dried Bone Custom-made Allografts for Maxillary Alveolar Horizontal Ridge Augmentation. Materials (Basel). 2020 Mar 19;13(6):1393. doi: 10.3390/ma13061393.

Frizziero L, Santi GM, Leon-Cardenas C, Ferretti P, Sali M, Gianese F, Crescentini N, Donnici G, Liverani A, Trisolino G, Zarantonello P, Stallone S, Di Gennaro GL. Heat Sterilization Effects on Polymeric, FDM-Optimized Orthopedic Cutting Guide for Surgical Procedures. J Funct Biomater. 2021 Nov 19;12(4):63. doi: 10.3390/jfb12040063.

Frizziero L, Santi GM, Leon-Cardenas C, Donnici G, Liverani A, Papaleo P, Napolitano F, Pagliari C, Di Gennaro GL, Stallone S, Stilli S, Trisolino G, Zarantonello P. In-House, Fast FDM Prototyping of a Custom Cutting Guide for a Lower-Risk Pediatric Femoral Osteotomy. Bioengineering (Basel). 2021 May 26;8(6):71. doi: 10.3390/bioengineering8060071.

Ferretti P, Leon-Cardenas C, Santi GM, Sali M, Ciotti E, Frizziero L, Donnici G, Liverani A. Relationship between FDM 3D Printing Parameters Study: Parameter Optimization for Lower Defects. Polymers (Basel). 2021 Jun 30;13(13):2190. doi: 10.3390/polym13132190.

Frizziero L, Pagliari C, Donnici G, Liverani A, Santi GM, Papaleo P, Napolitano F, Leon-Cardenas C, Trisolino G, Zarantonello P, Di Gennaro GL, Stilli S, Stallone S. Effectiveness Assessment of CAD Simulation in Complex Orthopedic Surgery Practices. Symmetry. 2021; 13(5):850.

Frizziero L, Santi GM, Liverani A, Napolitano F, Papaleo P, Maredi E, Gennaro GLD, Zarantonello P, Stallone S, Stilli S, Trisolino G. Computer-Aided Surgical Simulation for Correcting Complex Limb Deformities in Children. Applied Sciences. 2020; 10(15):5181.

Osti F, Santi GM, Neri M, Liverani A, Frizziero L, Stilli S, Maredi E, Zarantonello P, Gallone G, Stallone S, Trisolino G. CT Conversion Workflow for Intraoperative Usage of Bony Models: From DICOM Data to 3D Printed Models. Applied Sciences. 2019; 9(4):708.

Frizziero L, Santi GM, Liverani A, Giuseppetti V, Trisolino G, Maredi E, Stilli S. Paediatric Orthopaedic Surgery with 3D Printing: Improvements and Cost Reduction. Symmetry. 2019; 11(10):1317.

Frizziero L, Santi GM, Leon-Cardenas C, Donnici G, Liverani A, Napolitano F, Papaleo P, Pagliari C, Antonioli D, Stallone S, Di Gennaro GL, Trisolino G, Zarantonello P. An Innovative and Cost-Advantage CAD Solution for Cubitus Varus Surgical Planning in Children. Applied Sciences. 2021; 11(9):4057.

Kim HJ, Yoo SY, Jeon TY, Kim JH. Model-based iterative reconstruction in ultra-low-dose pediatric chest CT: comparison with adaptive statistical iterative reconstruction. Clin Imaging. 2016 Sep-Oct;40(5):1018-22. doi: 10.1016/j.clinimag.2016.06.006. Epub 2016 Jun 16.

Chiesa AM, Spinnato P, Miceli M, Facchini G. Radiologic Assessment of Osteosarcoma Lung Metastases: State of the Art and Recent Advances. Cells. 2021 Mar 4;10(3):553. doi: 10.3390/cells10030553.

Meyer S, Hirsch JM, Leiggener CS, Msallem B, Sigron GR, Kunz C, Thieringer FM. Fibula Graft Cutting Devices: Are 3D-Printed Cutting Guides More Precise than a Universal, Reusable Osteotomy Jig? J Clin Med. 2020 Dec 20;9(12):4119. doi: 10.3390/jcm9124119.

Zelen M. The randomization and stratification of patients to clinical trials. J Chronic Dis. 1974 Sep;27(7-8):365-75. doi: 10.1016/0021-9681(74)90015-0. No abstract available.

Thibault, JB. Veo: CT Model-Based Iterative Reconstruction. 2010. Available online: http://www.gehealthcare.com