Page 383 - IJB-10-2

International Journal of Bioprinting AM evaluation of medical device companies

more likely to initiate a crack than those presented in the printing technology (SLM or EBM), different machines,
center of the component. P4 and P5 had no defects on printing parameters, approaches to orientate the parts in the
20
the surface and low presence of voids. printing built, postprocessing strategies, communication
P8 presented a visible layer of alpha-case which showed approaches, etc. We even found variability in the regulatory
oxidation and corrosion. This layer could cause a negative pathway of the same implant. C3, C4, and C6 commercialize
effect on mechanical properties, such as ductility, fracture this type of pelvic implant as a class III medical device, while
toughness, and fatigue life, of the part especially under the remaining companies classified it as a class IIB device.
dynamic loading. 26 Therefore, manufacturers demonstrated different workflows
to produce the same implant. Due to these variables, the
All the prototypes in this study, except for P5, same implant design implied fabrication problems to some
demonstrated an acicular morphology with a dendritic companies and no issues to others. Furthermore, these
distribution. The presence of columnar grains implies that variations also produced parts with different microstructures
the relationship between grain orientation and mechanical and therefore, mechanical properties. Additionally, this
anisotropy must be considered, especially for a pelvic implant, study has proved that when the machine settings, process
subjected to high fatigue loading. In particular, P2, P6, and parameters, and postprocesses are not properly chosen, the
P8 showed more heterogeneity and thinner grains, which resulting parts can show imperfections that could affect the
are indicators of an inadequate thermal management during integrity of the components.
production and least favorable mechanical responses. 26
The manufacturing specifications were defined as
Only P5 presented a homogeneous and globular general requirements to be discussed with the companies,
structure with equiaxed and small grains, displaying and they generated a variety of opinions. Firstly, there was
the best microstructure in this study that theoretically a debate about which AM technology, SLM or EBM, is best
leads to the longest fatigue life. This is highly likely the for implant manufacture. These alternative technologies
27
consequence of applying the optimal thermal processes create distinct microstructures with different mechanical
that achieved appropriate recrystallization from the typical properties. Secondly, the need to perform thermal
28
acicular microstructure of PBF technology. 18 postprocesses such as HIP was controversial, potentially
because adding an HIP procedure considerably increases
3.2.4. Communication the cost and time for development. Only two companies,
The way companies managed communication during C5 and C6, performed HIP. C5 always incorporates an HIP
manufacturing was diverse. Most companies did not process to manufacture orthopedic implants. This company
provide any feedback about the design of the part, not provided a complete metallurgy analysis that demonstrated
promoting DfAM practices and manufactured the STL how the HIP process enhanced the microstructure of the part
file like a “service bureau” without adapting production and improved its fatigue strength. The literature supports
process to the design or function of the component. C4 this opinion for the development of parts that are subject
and C5 adapted the production process to the function of to high fatigue loading. C6, on the other hand, does not
20
the part, providing feedback about the design to optimize apply HIP to their implants but performed it for this study.
manufacturing and results. These two companies were the C4 applies HIP processes to some implants, but not usually
ones that provided P4 and P5, the best prototypes in terms to pelvic implants; therefore, they did not perform it for this
of geometry and microstructural integrity.
prototype. The rest of companies believed HIP process is not
4. Discussion required for the purpose of this pelvic component.
All the manufacturers knew that the parts were going
This study has provided the insights and trends on how to be evaluated. However, several imperfections were
some metal additive manufacturers produce custom identified in five out of the eight prototypes received. Further
medical devices in Europe. The analysis has been performed analyses are required to determine the actual clinical and
from a neutral perspective as none of the authors have any mechanical consequences of such imperfections. In the
conflicts of interest with the companies, ensuring that the visual inspection analysis, only three prototypes P4, P5,
results are objective and reliable. and P6 did not show any inconsistencies. In the geometrical
Our results show that the current lack of standardization precision evaluation, the same P4, P5, and P6 showed the
in the AM of metallic implants presents several challenges lowest values of geometrical deviations with the maximum
for the development of consistent and high-quality devices. values located in small areas of lattice structures, which
It has also illustrated the variability among the companies could be neglected. The other five prototypes, P1, P2, P3,
when managing the fabrication of the same implant design. P7, and P8, presented large maximum deviations of 3.00,
Each company had a tendency toward a preferred 3D 2.53, 4.94, 4.88, and 3.60 mm, respectively, that could have

Volume 10 Issue 2 (2024) 375 doi: 10.36922/ijb.0140