Applying a trauma custom bridge system to reconstruct critical-sized femoral and tibial defects: Finite element analysis, osseointegration, and clinical outcomes

To the Editor: Bone is a complex tissue that continually undergoes dynamic biological remodeling. However, when bone defects exceed the self-repair threshold, they are regarded as critical-sized bone defects (CBDs),[1] and artificial interventions are required to assist the repair process. Implanting three-dimensional (3D)-printed prostheses is a promising candidate for reconstructing limb CBDs. The geometry, pore size, and porosity of these printed prostheses can be customized via a layer-by-layer manufacturing process based on patient-specific model data.[2] Many studies have been conducted to introduce 3D-printed prostheses for repairing CBDs, confirming their biocompatibility, stability, and osteogenesis. However, systematic clinical reports of serial cases are still lacking so far. Our team independently developed a Trauma Custom Bridge (TCBridge) system specifically for the efficient repair of limb CBDs. As displayed in Supplementary Figure 1, https://links.lww.com/CM9/C94, this system consists of 3D-printed osteotomy guide plates, bone cement molds, skeleton models, and porous Ti6Al4V (Ti6Al4V powder, Beijing AKEC Medical Co., Ltd, Beijing, China) prostheses. The types of prostheses mainly included wingless prosthesis, single-wing prosthesis, and integrated nail-wing prosthesis. In this study, we first elucidated the dynamic characteristics of stress distribution via finite element analysis (FEA) models. Then we comprehensively collected and compared the different clinical effects based on serial cases. The detailed outcomes described here represent a systematic case series to report the use of 3D-printed prostheses to repair femoral and tibial CBDs. Stress, strain, and displacement distributions were simulated and calculated by means of HyperMesh 14.0 (Altair, Michigan, USA) software and FEBio 3.7 software (https://febio.org/) under specific physiological loading. The entire lower limb was used as the analytical model. All meshes were tetrahedral having a size of 1 mm. The details of the five models can be found in Supplementary Figure 2, https://links.lww.com/CM9/C94. The properties of the different materials are listed in Supplementary Table 1, https://links.lww.com/CM9/C94. The characteristics of stress conduction were analyzed by simulating the step movement. Before step, the reaction force values were X-axis = 90.2 N, Y-axis = 729.1 N, and Z-axis = 36.1 N; after step, the reaction force values were X-axis = 63.8 N, Y-axis = 797.8 N, and Z-axis = 57.6 N. Then, the accuracy verification of the FEA models was finished. This clinical study was approved by the Medical Science Research Ethics Committee of Peking University Third Hospital (No. M2018174). Informed consents were obtained from all patients and their families. The inclusion criteria were (1) CBDs caused by osteomyelitis and aseptic non-union; (2) length >6 cm; and (3) age >18 years. The exclusion criteria were (1) tumor; (2) CBDs involving articular surface; and (3) patients unable to fulfill the entire process. All enrolled patients were divided into Femoral and Tibial CBDs groups. Customized bone cement molds, osteotomy guide plates, and prostheses were all designed using the Mimics 22.0 software (Materialize, Leuven, Belgium). Polyamide powders were used to 3D-print the molds and guide plates (3D printer: EOS FORMIGA P110, Germany). Ti6Al4V powders were used to 3D-print the prostheses (3D printer: ARCAM EBM Q10, Goteborg, Sweden). The surgical process comprised two main stages: (1) thorough debridement and spacer insertion, and (2) prosthesis implantation and internal fixation. After surgery, gradual full weight-bearing was mainly determined based on radiological follow-up results. The Lower Extremity Functional Scale (LEFS) was used to evaluate limb function quantitatively. SPSS 22.0 software (IBM Software, Armonk, NY, USA) was used to execute the statistical analysis. Variable data were presented as means ± standard deviations and assessed by the Kruskal–Wallis test. Attribute data were presented as numbers (percentages) and assessed by the χ2 test. Statistical significance was considered at P-value <0.05. As shown in Supplementary Figures 3 and 4, https://links.lww.com/CM9/C94, stress mainly concentrated on the proximal areas of bones and the marginal ends of prostheses. The details about quantitative von mises, strain and relative displacement can be found in Supplementary Figure 5, https://links.lww.com/CM9/C94. As shown in Supplementary Table 2, https://links.lww.com/CM9/C94, 32 patients (33 defects) were enrolled. The average length and percentage of defects on the bones were 13.6 cm and 35.1%. The etiological factors included osteomyelitis (24 defects) and aseptic non-union (9 defects). After prostheses implantation, the lengths of defect-located femurs and tibias increased to 42.0 cm and 37.0 cm. The average beginning times of weight-bearing and follow-up periods were 10.5 days and 32.4 months. All patients’ preoperative LEFS scores were 27.2, which significantly improved to 53.6 at the last follow-up (P <0.001). Regarding the postoperative complications, one case (1/24, 4.2%) had osteomyelitis recurrence, six (6/33, 18.2%) experienced screw loosening or breakage, and two (2/33, 6.1%) experienced intramedullary nail breakage because of acute trauma. As shown in Figure 1 and Supplementary Figure 6, https://links.lww.com/CM9/C94, the growth of new bone began with the local proliferation and formed a “pedestal-like” structure, and then crawled along the prosthesis. Besides, we obtained an implanted prosthesis from a recurrent tibial osteomyelitis and performed a Hematoxylin and Eosin and Toluidine Blue staining, which jointly revealed that new bone had grown inside the prosthesis pores and formed stable osseointegration [Supplementary Figure 7, https://links.lww.com/CM9/C94].Figure 1: Typical case of femoral osteomyelitis-derived defect: (A1–F1) anteroposterior X-rays at 1, 7, 20, 35, 47, 61 months after prosthesis implantation; (A2–F2) lateral X-rays at 1, 7, 20, 35, 47, 61 months after prosthesis implantation. Yellow arrows directed the newly-regenerated bone, and white arrows indicated the 3D-printed prostheses.The TCBridge system can provide an entire personalized reconstruction strategy. Prostheses with interconnected pores can mimic the extracellular matrix of natural bone and provide a conducive microenvironment to promote vascularization and nutrient exchange. The customized guide plate can ensure an accurate osteotomy. The customized bone cement mold allows the spacer to have an anatomical shape. From the view of FEA, the stress on the prostheses was mainly concentrated on the integrated nail and lateral wing, and marginal ends. Adding integrated nail and wing structures can alter the stress conduction, which improves the prosthesis stability. Combining the previous and present results, we systematically described the stress distribution characteristics of using 3D-printed prostheses to reconstruct limb CBDs, providing significant references for further prosthesis design and clinical practice. In this study, the average beginning time of limb weight-bearing was 10.5 days and presented no significant difference between the two groups. By comparison, a previous meta-analysis showed that the times to full weight-bearing in the Masquelet and Ilizarov groups were 7.67 months and 9.45 months, respectively.[3] After prosthesis implantation, the anatomical bone integrity could be reconstructed immediately, and limb stability and stress conduction could also be restored, thereby ensuring limb weight-bearing. The sustained stability of a prosthesis was crucial for postoperative weight-bearing. The whole stabilization process of using 3D-printed prostheses could be divided into primary and long-term stability. Primary stability mainly relied on the friction between the rough surfaces and the mechanical interlocking derived from internal fixation. The long-term stability process was synchronized with continuous recovery. The newly regenerated bone and soft tissues could grow inside the prosthesis to form mechanical anchoring. The process of new bone regeneration followed the “secondary fracture healing” pattern.[4] The early micro-motion between the bone and prosthesis stimulated local callus hyperplasia. Then new bone gradually crept along the prosthesis and remodeled more compliantly and smoothly. The remodeling process was regulated by stress conduction, which conformed to Wolff’s law. It is important to specifically emphasize that we did not implant additional bone materials. Relying on the porous structure of the prosthesis and the appropriate stability pattern, new bone could constantly grow and osseointegrate at the interfacial sites between the host bone and prosthesis. The induced membrane around the prosthesis also played an important promoting role during the bone-regeneration stage. Regarding to the functional recovery, the LEFS scores of all patients showed significant improvement at the last follow-up than before surgery (P <0.001). All patients could walk independently and return to normal activities. Accurate reconstruction of bone anatomical structure and stress conduction, stable fixation, early and safe weight-bearing exercises, and continuous bone regeneration were important guarantees for patients to achieve long-term limb function recovery. Although a lot of work has been done, there are some limitations in this study. First, the elastic modulus mismatch between the Ti6Al4V and bone may lead to stress shielding and potential implant failure. Second, the uniform porous structure of prosthesis makes it difficult to accurately reconstruct the biomechanical and structural microenvironment. Our further research will focus on manufacturing anisotropic 3D-printed prostheses to achieve precise reconstruction. In conclusion, the TCBridge system exhibited excellent clinical effectiveness in repairing femoral and tibial CBDs. Patients’ bone shortening could be significantly improved, achieving early weight-bearing and functional exercise, and new bone continued to regenerate and osseointegrate with the prosthesis, ultimately achieving better limb function. Funding This clinical study was supported by the grants from Beijing Municipal Science & Technology Commission (No. Z181100001718195), National Natural Science Foundation of China (No. 82172065), Key Clinical Projects of Peking University Third Hospital (No. BYSYZD2022031), and Peking University Medicine Sailing Program for Young Scholars’ Scientific & Technological Innovation (No. BMU2023YFJHPY015). Conflicts of interest None.