Patient-Specific 3-Dimensional Printing Titanium Implant Biomechanical Evaluation for Complex Distal Femoral Open Fracture Reconstruction with Segmental Large Bone Defect: A Nonlinear Finite Element Analysis
Round 1
Reviewer 1 Report
This is a well written paper, however i have some general and specific comments.
In general, since what is being proposed is a method that is patient-specific, you will need more tests and evidence in order to confirm the viability of the method. One patient with one set of FE results is a good starting point, but I do recommend that you subsidize this one study with other similar ones. It is understandable that similar injuries cannot be replicated, however other fracture comminutions that have undergone equivalent FE-based prosthetic analyses may be helpful.
Also, this mentions that the method/phenomenon can be greatly improved by rounding the angles of the implant lattice structure. Have you showed that? If so, please show the results. Explain how will rounding the lattice angles affect the inter-elemental mesh nodal degrees of freedom in the FE model.
Specific comments:
L. 44: define 3DP as 3D Printing.
L. 64: please change the abbreviation PSI to PsI or other. The former is usually reserved to express Pound-force per Square Inch.
L. 85: A 17 y/o male who...
Fig 1: No need to show the actual bloody fracture. Please remove it. The x-ray suffices.
L. 112: please use x for multiplication, not *.
Fig 2: (b) is missing in the bottom figure.
L. 121: please remove yellow highlights.
L. 142: please remove dot "." in front of 3. Results.
L. 149: at the junction with the bone on bone.. or bone to bone.
Fig 4: please add label (with units) for y-axis.
L. 171: remove "The" in front of "stage 3".
L. 176: biocompatibility with regard to size.
L. 184: please explain what you mean by "ultrasonic oscillations".
L. 193: please put ref number after author's name: Frost [16]
L. 194: please put ref number after authors' name: Rubin and Lanyon [17]
L. 219: was considered as the loading condition.
L. 226: This portion of the Conclusions is too abrupt and insinuating that the rounded structure has been analyzed; which is not supported by evidence. Please modify this section accordingly.
Author Response
General Comments
Comment 1:
This is a well written paper, however i have some general and specific comments.
In general, since what is being proposed is a method that is patient-specific, you will need more tests and evidence in order to confirm the viability of the method. One patient with one set of FE results is a good starting point, but I do recommend that you subsidize this one study with other similar ones. It is understandable that similar injuries cannot be replicated, however other fracture comminutions that have undergone equivalent FE-based prosthetic analyses may be helpful.
Response 1:
As mentioned in the article, distal femur fractures with large bone defect are not very common, but are very serious injuries. Treatment of this type of injury is complicated and difficult. Traditional surgical methods cannot provide benefits such as those achieved by 3D printing implants, such as complete filling of defects and providing tight fracture contact, which can further achieve good surgical stability.
However, the use of novel 3D printing implant requires rigorous experimental verification before clinical application. Novel design such as the proximal stem support, the upper and lower shell with porous structure, the external lattice structure etc. are clinically innovative designs. In this study, we hope to verify the feasibility of biomechanics through FE analysis. At the same time, the implant is also actual produced through 3D printing implementation. With good biomechanics performance and production feasibility, we plan to arrange in vitro biomechanical experiment and clinical trial for future work.
In the article ‘Putting 3D modelling and 3D printing into practice: virtual surgery and preoperative planning to reconstruct complex post-traumatic skeletal deformities and defects’ . SICOT J. 2017;3:16, by Tetsworth K et al, they successfully treated similar fractures with a 3D printed truss implant. To some extent, the 3D printed truss implant they designed should be more precisely called a metal spacer or a cage. Although it has been successfully used in clinical treatment of patient, but there is a lack of evidence related to biomechanical research. In contrast, in this study we hope that the ideal implant can be designed, then tested and optimized, before entering to the clinical application.
Comment 2:
Also, this mentions that the method/phenomenon can be greatly improved by rounding the angles of the implant lattice structure. Have you showed that? If so, please show the results. Explain how will rounding the lattice angles affect the inter-elemental mesh nodal degrees of freedom in the FE model.
Response 2:
To decrease the high stress at the sharp area at the lattice structure, original solid model (CAD model) was modified to rounded angle at the sharp areas and re-meshed using smart automatically mesh procedure provided in ANSYS. The design of sharp and modified rounded angle lattice structure is shown in the Figure 9. The result of the stress distribution is showed in the Figure 4, where rounded angle group is represented by slash blue bar. After the rounded angle design is introduced, the stress dropped from 1318 MPa to nearly 600 MPa. We think this is reasonable, because the stress can be distributed more evenly through the round smooth margin of the lattice structure. Thus, concentration of the stress will be remarkably reduced.
Specific comments:
C 1: L. 44: define 3DP as 3D Printing.
R 1: Revised as requested.
C 2: L. 64: please change the abbreviation PSI to PsI or other. The former is usually reserved to express Pound-force per Square Inch.
R 2: Revised as requested, we agree to use the abbreviation PsI for patient specific implant.
C 3: L. 85: A 17 y/o male who...
R 3: Revised as requested.
C 4: Fig 1: No need to show the actual bloody fracture. Please remove it. The x-ray suffices.
R 4: The clinical picture was removed.
C 5: L. 112: please use x for multiplication, not *.
R 5: Revised as requested.
C 6: Fig 2: (b) is missing in the bottom figure.
R 6: Revised as requested.
C 7: L. 121: please remove yellow highlights.
R 7: Revised as requested.
C 8: L. 142: please remove dot "." in front of 3. Results.
R 8: Revised as requested.
C 9: L. 149: at the junction with the bone on bone.. or bone to bone.
R 9: Revised as requested.
C 10: Fig 4: please add label (with units) for y-axis.
R 10: Revised as requested.
C 11: L. 171: remove "The" in front of "stage 3".
R 11: Revised as requested.
C 12: L. 176: biocompatibility with regard to size.
R 12: Revised as requested.
C 13: L. 184: please explain what you mean by "ultrasonic oscillations".
R 13: Ultrasonic oscillating washing device is used to wash away the metal particle from the implant, it is an effective tool for cleaning the porous structure of the implant.
C 14: L. 193: please put ref number after author's name: Frost [16]
R 14: Revised as requested.
C 15: L. 194: please put ref number after authors' name: Rubin and Lanyon [17]
R 15: Revised as requested.
C 16: L. 219: was considered as the loading condition.
R 16: Revised as requested.
C 17: L. 226: This portion of the Conclusions is too abrupt and insinuating that the rounded structure has been analyzed; which is not supported by evidence. Please modify this section accordingly.
R 17: The conclusion is modified as below:
The proposed novel patient-specific 3D Printing implant design concept for a large distal femoral fracture defect can fulfill the clinical need of optimum restoration of the injured limb. The implant with rounded angle lattice structure provides excellent biomechanical environment under a single leg standing phase which may promote fracture healing.
Reviewer 2 Report
The paper develops a patient-specific implant for a bone defects and validates the design using finite element analysis. The design and analysis are carefully conducted but the paper lacks generality to other applications and replication for further research groups and full support of conclusions. For instance, which aspects of this study are specific to the patient and which are generalizable conclusions and approaches that further groups could use? There is also a need for better comparisons of study decisions and results with other works.
A primary concern in the paper is whether the authors considered tailoring the pore size for improved osseointegration or considered polymer materials that could reduce stress shielding as seen in recent bone tissue engineering works applicable to patient-specific implants:
Egan, P. F., Shea, K. A., & Ferguson, S. J. (2018). Simulated tissue growth for 3D printed scaffolds. Biomechanics and modeling in mechanobiology, 17(5), 1481-1495.
Kang, H., Hollister, S. J., La Marca, F., Park, P., & Lin, C. Y. (2013). Porous biodegradable lumbar interbody fusion cage design and fabrication using integrated global-local topology optimization with laser sintering. Journal of biomechanical engineering, 135(10).
Comments are as follows:
P3 line 98 How many models were constructed, were these from only one patient? What aspects of this approach are generalizable to further patients?
P5 What is the standard length of time expected for the patient to be in stage 1, 2, and 3 of healing? What conditions dictate whether the design is successful at each stage?
P6 At the beginning of the results, the preferred outcome of the design approach is not stated, is reduction of stress the primary objective? Is osseointegration capability evaluated?
Figure 4 has no y-axis label/units, and only one figure for results seems a bit short. Much of the discussion could be moved in results/earlier sections especially concerning how stage 1/2/3 differ to give context to earlier decisions.
P9 What is the optimization criteria for the structure?
Figure 9 and surrounding discussion is vague and it seems the stress reduction is minor and would not significantly improve the performance of the structure. The authors state an around 600 MPa stress but should use specific values when comparing.
The discussion should have more direct comparisons to other studies and alternate design approaches for patient-specific tailoring and materials for bone tissue engineering such as those in references above.
Author Response
General Comments
Comment 1:
The paper develops a patient-specific implant for a bone defects and validates the design using finite element analysis. The design and analysis are carefully conducted but the paper lacks generality to other applications and replication for further research groups and full support of conclusions. For instance, which aspects of this study are specific to the patient and which are generalizable conclusions and approaches that further groups could use? There is also a need for better comparisons of study decisions and results with other works.
Response 1:
As mentioned in the article, distal femur fractures with large bone defect are not very common, but are very serious injuries. Treatment of this type of injury is complicated and difficult. Traditional surgical methods cannot provide benefits those achieved by 3D printed implants, such as complete filling of defects and providing tight fracture contact, which can further achieve good surgical stability.
We think that the use of novel 3D printing implant requires rigorous experimental verification before clinical application. Novel design such as the proximal stem support, the upper and lower shell with porous structure, the external lattice structure etc. are clinically innovative designs. In this study, we hope to verify the feasibility of biomechanics through FE analysis. At the same time, the implant is also actual produced through 3D printing implementation. With good biomechanics performance and production feasibility, we plan to arrange in vitro biomechanical experiment and clinical trial for future work.
In article ‘Putting 3D modelling and 3D printing into practice: virtual surgery and preoperative planning to reconstruct complex post-traumatic skeletal deformities and defects’ . SICOT J. 2017;3:16, by Tetsworth K et al, they successfully treated similar fractures with a 3D printed truss implant. To some extent, the 3D printed truss implant designed in the article should be more precisely called a metal spacer or a cage. Although it has been successfully used in clinical treatment of patient, but there is a lack of evidence related to biomechanical research. In contrast, in this study we hope that the ideal implant can be designed, then tested and optimized, before entering to the clinical application.
This 3D printing implant is designed specifically for the patient we mentioned in the article. The size, the shape, the contour and the morphology of the implant is tailored to fit the fracture defect. However, we think there are several design concepts can be reference for application extension:
- Proximal and distal porous shell design. This design can provide good support by filling up all of the fracture area and it can minimize the dead space between the bone-implant interface, Also, the porous structure act as a scaffold for bone integration. This design can be applied into the implant design of juxta articular fractures with bone defect such as the distal tibia fracture.
- The protruding stem design on the proximal shell. This design can provide additional stability when facing meta-diaphyseal fracture. For instance, if the fracture defect is on the middle part of the femur shaft, protruding stem can be designed on each end of the implant, which facilitate to be fit into the femoral canal for fixation.
- The lattice structure with rounded angle. As the body of the implant, this structure can minimize stress concentration as were showed in the study. It can ensure the longevity of the implant and also act as a space storage for bone graft.
Comment 2:
A primary concern in the paper is whether the authors considered tailoring the pore size for improved osseointegration or considered polymer materials that could reduce stress shielding as seen in recent bone tissue engineering works applicable to patient-specific implants:
Egan, P. F., Shea, K. A., & Ferguson, S. J. (2018). Simulated tissue growth for 3D printed scaffolds. Biomechanics and modeling in mechanobiology, 17(5), 1481-1495.
Kang, H., Hollister, S. J., La Marca, F., Park, P., & Lin, C. Y. (2013). Porous biodegradable lumbar interbody fusion cage design and fabrication using integrated global-local topology optimization with laser sintering. Journal of biomechanical engineering, 135(10).
Response 2:
Thank you for your practical comment. The purpose of this study is to design an optimum 3D printed implant to treat patient of distal femur fracture with large bone defect using “general non-linear finite element analysis with global model”. The implant/bone interface was only set as two different conditions in global model: contact (bone integration is not complete, as in stage 1) or bonded (bone integration is complete, as in stage 2 and 3). The global model of finite elements analysis is not suitable for investigating the growth of the bone within the micro-porous structure, and the size of the pores. The computer simulation analysis of the detailed interface mechanics can only be achieved through multi-scale or submodeling analysis methods. However, this is beyond the purpose of this study. Therefore, above two articles might not suitable as the references for our article.
Specific comments:
Comment 1:
P3 line 98 How many models were constructed, were these from only one patient? What aspects of this approach are generalizable to further patients?
Response 1:
The model was obtained and constructed from both femurs CT scans from one patient. But in order to explore three different stages, we have established 3 models accordingly. The digital approach converting CT scan of the bone to 3D solid model by using CAD system is a generalizable method for further patients. By using mirror technique, the injured and the normal limbs can be overlapped. Therefore, deformity such as injured limb shortening, rotation, angulation etc. can be corrected. After the corrections, the fracture defect with clear margin can be defined. Also, the fracture defect size and volume can be estimated. Model making and analysis procedures can be standardized and optimized, which can be applied for further patients.
Comment 2:
P5 What is the standard length of time expected for the patient to be in stage 1, 2, and 3 of healing? What conditions dictate whether the design is successful at each stage?
Response 2:
For normal bone healing process, callus formation is expected at 3 months postoperatively, while fracture complete union will be achieved at 1-2 yrs. But for bone remodeling and consolidation, it maybe takes another 1-2 years after fracture complete union. Therefore, Stage 1 is the immediate postoperative stage, the expected time is approximately 1-2 years; Stage 2 is the stage when fracture union is complete, for this stage it is expected to take 1-2 years before complete fracture consolidation; For Stage 3 is the stage of removal of locking plate after fracture consolidation.
Comment 3:
P6 At the beginning of the results, the preferred outcome of the design approach is not stated, is reduction of stress the primary objective? Is osseointegration capability evaluated?
Response 3:
The objective of this study is to propose a novel “gobal” design with good biomechanical performance. In the result part, we showed significant reduction of stresses from stage 1 to stage 2. This result ensures the feasibility and longevity of the implant and good environment for fracture union. This is important because implant failure and loss of reduction normally occur in stage 1. In clinical practice, reaching stage 2 with complete osteointegration (interface is bonded) can allow patients to have acceptable outcome. A stable and functional limb can be achieved. Osteointegration evaluation is out of the scope of this study. As we mentioned before, global model of finite element analysis is not suitable for detailed interface osteointegration evaluation.
Comment 4:
Figure 4 has no y-axis label/units, and only one figure for results seems a bit short. Much of the discussion could be moved in results/earlier sections especially concerning how stage 1/2/3 differ to give context to earlier decisions.
Response 4:
Y-axis label/units is added to figure 4. Figure 4, 5, 6 and 7 are included in the results. The discussion is adjusted accordingly.
Comment 5:
P9 What is the optimization criteria for the structure?
Response 5:
We think that an optimum implant should have criterias as below
- Can replace the bone defect as it is defined. And can closely fit to the fracture area with minimal dead space
- Structure strong enough to provide good initial stability. The cylinder tube, the lattice structure and the proximal/distal porous shells will ensue the feasibility and longevity of the implant.
- As light weight as possible. The weight of the implant is only 1.37 times heavier than the original bone (implant sold weight is 107 gm; the corresponding defect bone is 78 gm).
- Easy to apply for the surgeon during the surgery. The design of the proximal protruding stem can be easily applied to the femur shaft, initial stability can be achieved once it fit into the canal. The distal porous shell which is closely match with the fracture site can also be fit into the defect fast and easy.
Comment 6:
Figure 9 and surrounding discussion is vague and it seems the stress reduction is minor and would not significantly improve the performance of the structure. The authors state an around 600 MPa stress but should use specific values when comparing.
Response 6:
The initial design with acute angle in the lattice structure, the stress value is as high as 1318 MPa (showed in Figure 4). After the design is modified to rounded angle, the stress reduced to around 600MPa, nearly decrease more than half of the initial value. Also, the later value is much lower when comparing to the titanium alloy fracture strength (1000MPa). Therefore, we think that this design can greatly improve the implant performance.
Comment 7:
The discussion should have more direct comparisons to other studies and alternate design approaches for patient-specific tailoring and materials for bone tissue engineering such as those in references above.
Response 7:
As response in general comments, in the article ‘Putting 3D modelling and 3D printing into practice: virtual surgery and preoperative planning to reconstruct complex post-traumatic skeletal deformities and defects’ . SICOT J. 2017;3:16, by Tetsworth K et al, they successfully treated similar fractures with a 3D printed truss implant. To some extent, the 3D printed truss implant they designed should be more precisely called a metal spacer or a cage. Although it has been successfully used in clinical treatment of patient, but there is a lack of evidence related to biomechanical research. In contrast, in this study we hope that the ideal implant can be designed, then tested and optimized, before entering to the clinical application.
Reviewer 3 Report
I believe its a very interesting paper in his field with a good presentation, good English, adequate design and scientifically interesting
Author Response
General comments:
I believe its a very interesting paper in his field with a good presentation, good English, adequate design and scientifically interesting
Response:
Thanks for the Reviewer’s encouragement.
Round 2
Reviewer 1 Report
Thank you for making the appropriate changes as requested.
Good job!
Author Response
Thank you
Reviewer 2 Report
The authors are encouraged to resubmit the manuscript and response to comments with major revisions to better address the original comments.
Much of the original comments have not been changed in the manuscript or are not indicated as changed in a sufficiently clear manner to determine if they have been addressed. In writing a response, it is expected that the authors write their response to the comment, indicate where/how the manuscript was changed in the rebuttal document, and then highlight the changes in red text in the document. Currently they do not indicate where/how the manuscript was changed and also do not seem to change it according to some of their responses (or are missing the red text to indicate all changes).
In regards to specific comments and how they have been address:
General comments:
Comment 1: Much of the author’s comment focuses restating what is already in the paper and steps for taking to clinical trials, rather than generalizability of the conclusions of scientific interest.
The authors state three design concepts that can be a reference but do not seem to have included this in the paper.
Comment 2:
The authors have not addressed how they choose the microstructure pore size which is relevant to ensure osseointegration, even if it isn’t explicitly modeled in their simulation. Therefore, there is a need to reference some other type of literature to justify these design decisions to ensure their goal of regenerating bone is accomplished.
Specific comments:
Comment 1:
The comment is addressed but the authors have not clearly indicated any changes in the paper which is necessary.
Comment 2:
The comment is addressed but the authors have not clearly indicated any changes in the paper which is necessary.
Comment 3:
The comment is partially addressed but the authors have not clearly indicated any changes in the paper which is necessary. The authors need to consider osseointegration in reference to existing sources, they do not need to conduct new analysis/simulations but do need to justify their decisions to show relevance that these criteria have been considered and can be explored in future studies.
Comment 4:
The authors have done minimal changes to address the comment in adjusting the discussion, most of their results still appear in the discussion section.
Comment 5:
The comment is addressed but the authors have not indicated any changes in the paper which is necessary.
Comment 6:
The authors have responded to the comment but not seem to have improved the paper to more clearly point out the significance.
Comment 7:
The authors have addressed the comment and it seems to be reflected in changes after Figure 9 in the manuscript. However, their writing is vague, the authors state “we hope” rather than providing definitive conclusions supported by the manuscript and also do not address patient-specific tailoring.
Author Response
Responses to Reviewers’ comments on manuscript: applsci-818523
Patient-Specific 3-Dimensional Printing Titanium Implant Biomechanical Evaluation for Complex Distal Femoral Open Fracture Reconstruction with Segmental Large Bone Defect: A Nonlinear Finite Element Analysis
Reviewer(s)' Comments to Author:
Much of the original comments have not been changed in the manuscript or are not indicated as changed in a sufficiently clear manner to determine if they have been addressed. In writing a response, it is expected that the authors write their response to the comment, indicate where/how the manuscript was changed in the rebuttal document, and then highlight the changes in red text in the document. Currently they do not indicate where/how the manuscript was changed and also do not seem to change it according to some of their responses (or are missing the red text to indicate all changes).
In regards to specific comments and how they have been address:
General comments:
Comment 1: Much of the author’s comment focuses restating what is already in the paper and steps for taking to clinical trials, rather than generalizability of the conclusions of scientific interest.
The authors state three design concepts that can be a reference but do not seem to have included this in the paper.
Response 1:
Changes were made in the manuscript in Line 249-264, regarding the generalizability of the implant designs and the three design concepts.
Comment 2:
The authors have not addressed how they choose the microstructure pore size which is relevant to ensure osseointegration, even if it isn’t explicitly modeled in their simulation. Therefore, there is a need to reference some other type of literature to justify these design decisions to ensure their goal of regenerating bone is accomplished.
Response 2:
Changes were made in the manuscript in Line 220-230.
The reason why we chose microstructure with 600 μm in physical 3D printed model is based on two articles listed below:
The first article stated that the development of high-porous titanium material with 600 μm shows high potential to be modern material for creating a 3D structure for bone regeneration and implant, because of its high permeability allows extensive body fluid transport through the porous implant. This can provoke bone tissue ingrowth.
The second article proved that when comparing different pore sizes (300μm, 600μm and 900μm), 600μm demonstrated higher fixation ability. Also, it has appropriate mechanical strength and rapid bone ingrowth.
- Bogdan D, Wojciech S, Dirk G. Highly porous titanium scaffolds for orthopaedic applications. J Biomed Mater Res B Appl Biomater.2010;95(1):53-61.
- Naoya T, Shunsuke F, MitsuruT. Effect of pore size on bone ingrowth into porous titanium implants fabricated by additive manufacturing: An in vivo experiment. Materials Science and Engineering: C. 2016;59(1): 690-701.
Specific comments:
Comment 1:
The comment is addressed but the authors have not clearly indicated any changes in the paper which is necessary.
(P3 line 98 How many models were constructed, were these from only one patient? What aspects of this approach are generalizable to further patients?)
Response 1:
Changes were made in the manuscript in Line 98, 118-120, 253-265.
Comment 2:
The comment is addressed but the authors have not clearly indicated any changes in the paper which is necessary.
(P5 What is the standard length of time expected for the patient to be in stage 1, 2, and 3 of healing? What conditions dictate whether the design is successful at each stage?)
Response 2:
Changes were made in the manuscript in Line 182-187.
Comment 3:
The comment is partially addressed but the authors have not clearly indicated any changes in the paper which is necessary. The authors need to consider osseointegration in reference to existing sources, they do not need to conduct new analysis/simulations but do need to justify their decisions to show relevance that these criteria have been considered and can be explored in future studies.
(P6 At the beginning of the results, the preferred outcome of the design approach is not stated, is reduction of stress the primary objective? Is osseointegration capability evaluated?)
Response 3:
Changes were made in the manuscript in Line 220-230.
Comment 4:
The authors have done minimal changes to address the comment in adjusting the discussion, most of their results still appear in the discussion section.
(Figure 4 has no y-axis label/units, and only one figure for results seems a bit short. Much of the discussion could be moved in results/earlier sections especially concerning how stage 1/2/3 differ to give context to earlier decisions.)
Response 4:
Some results in the discussion section were moved to the result section (Line 152-154, 158-166).
Comment 5:
The comment is addressed but the authors have not indicated any changes in the paper which is necessary.
(P9 What is the optimization criteria for the structure?)
Response 5:
The optimization criteria were added in the manuscript (Line 199-208).
Comment 6:
The authors have responded to the comment but not seem to have improved the paper to more clearly point out the significance.
(Figure 9 and surrounding discussion is vague and it seems the stress reduction is minor and would not significantly improve the performance of the structure. The authors state an around 600 MPa stress but should use specific values when comparing.)
Response 6:
Changes were made in the result section of the manuscript in Line 158-162.
Comment 7:
The authors have addressed the comment and it seems to be reflected in changes after Figure 9 in the manuscript. However, their writing is vague, the authors state “we hope” rather than providing definitive conclusions supported by the manuscript and also do not address patient-specific tailoring.
(The discussion should have more direct comparisons to other studies and alternate design approaches for patient-specific tailoring and materials for bone tissue engineering such as those in references above.)
Response 7:
Changes was made in the manuscript (Line 279-282).
Round 3
Reviewer 2 Report
The authors have revised the manuscript accordingly and it is suitable for publication after proofreading and grammar corrections.
