Comparative analysis of current 3D printed acetabular titanium implants
The design freedom allowed by three-dimensional (3D) printing enables the production of acetabular off-the-shelf cups with complex porous structures. The only studies on these designs are limited to clinical outcomes. Our aim was to analyse and compare the designs of different 3D printed cups from multiple manufacturers (Delta TT, Trident II Tritanium and Mpact 3D Metal).
We analysed the outer surface of the cups using scanning electron microscopy (SEM) and assessed clinically relevant morphometric features of the lattice structures using micro-computed tomography (micro-CT). Dimensions related to the cup wall (solid, lattice and overall thickness) were also measured. Roundness and roughness of the internal cup surface were analysed with coordinate measuring machine (CMM) and optical profilometry.
SEM showed partially molten titanium beads on all cups, significantly smaller on Trident II (27 μm vs ~ 70 μm, p < 0.0001). We found a spread of pore sizes, with median values of 0.521, 0.841 and 1.004 mm for Trident II, Delta TT and Mpact, respectively. Trident II was also significantly less porous (63%, p < 0.0001) than the others (Delta TT 72.3%, Mpact 76.4%), and showed the thinnest lattice region of the cup wall (1.038 mm, p < 0.0001), while Mpact exhibited the thicker solid region (4.880 mm, p < 0.0044). Similar roundness and roughness of the internal cup surfaces were found.
This was the first study to compare the designs of different 3D printed cups. A variability in the morphology of the outer surface of the cups and lattice structures was found. The existence of titanium beads on 3D printed parts is a known by-product of the manufacturing process; however, their prevalence on acetabular cups used in patients is an interesting finding, since these beads may potentially be released in the body.
Keywords3D printing Additive manufacturing Orthopaedic implant Acetabular cup Hip Arthroplasty
More than 90,000 total hip arthroplasties (THA) were performed in the United Kingdom in 2017, with almost 70% using uncemented acetabular cups . Although the majority of these implants were manufactured using traditional technologies [2, 3], around 10% of off-the-shelf designs are now produced using three-dimensional (3D) printing methods [4, 5].
The design freedom allowed by 3D printing enables the production of cups with different features and complex porous structures ; the main clinical rationale is to provide enhanced fixation with bone, compared to conventionally manufactured cups [7, 8, 9]. This is particularly pertinent given that aseptic loosening is one of the most common reason for revision in THA .
Over 60,000 acetabular cups have been produced using electron beam melting (EBM), a common 3D printing technique for metal components . These implants have undergone benchmark testing and clinical evaluation to obtain the required certifications for application [11, 12, 13], but the implications of 3D printing manufacturing on the properties and performance of these implants are yet to be published in the literature.
The only published studies to investigate off-the-shelf 3D printed acetabular components have been limited to early to medium term clinical outcomes [14, 15, 16] and there have been no studies to report on independent analysis methods for evaluating 3D printed implants.
The aim of this study was to compare different designs of titanium 3D printed off-the-shelf cups for THA. Our first objective was to investigate characteristics of the outer surface of the cups; our second objective was to assess features of the internal cup surface.
Materials and methods
Micro-computed tomography (micro-CT) was used to characterize the lattice structure on the outer surface of the cups. The analysis involved the following steps: (1) micro-CT data acquisition, (2) data reconstruction and segmentation, (3) measurement of morphometric parameters of the lattice structure and dimensions related to cup wall.
A micro-CT scanner (XTH 225, Nikon Metrology, UK) equipped with a Tungsten target as X-ray source and set at beam voltage and current of 150 kV and 70 μA, respectively, was used to scan the implants. A Cu (0.25 mm) hardware filter was used, located at the beam source, to reduce the beam hardening effect, which involves the attenuation of low energy photons (i.e. soft X-rays) while the X-rays beam moves through the absorbing material. All the scans were performed at a complete 360° rotation at a step size of 0.11°, with 3177 views, 1 frame per view and exposure time of 1000 ms. The images resolution was 33 μm, with a total scan time per sample of 53 min.
Data reconstruction and segmentation
A volume model of the scanned implants was reconstructed from the acquired 2D projection images importing the data into CT Pro 3D software (version 4.4, Nikon Metrology, UK), using a filtered back-projection algorithm. During the reconstruction process, a numerical filtering (polynomial correction of second order) was applied to further reduce the beam hardening effect.
The reconstructed images of the implants were segmented and rendered using the 3D micro-CT analysis software Avizo (version 9.0, Thermofisher Scientific, US). Segmentation was performed using a built-in automatic segmentation process based on the “iso-50%” principle, using a specific edge grey value for each implant. The threshold value corresponded with the mid-grey-level between the peaks that coincide with the irradiated materials (in our case Titanium alloy and air as background) in the histogram plot of voxels count versus voxel intensity.
Measurements of morphometric parameters
All the measurements were performed using Avizo and the public domain software for image analysis ImageJ (version 1.52a, Broken Symmetry Software). Regions of interest confined to the porous network only were selected for porosity, pore size and strut thickness; 15 measurements were made of each parameter on each component. A number of 8 equidistant cross-sections passing for the pole of the cups were selected, taking 5 measurements of solid, lattice and overall thickness on each cross-section for each component.
Scanning Electron microscopy (SEM)
Detailed analysis of the morphology of the outer surface of the implants was performed using a scanning electron microscopy - SEM (Hitachi S3400-N, Tokyo, Japan). Images were captured in secondary electron imaging (SEI) mode at 20 kV, with magnifications ranging from 20x to 250x.
The roundness of the internal cup surface was measured using a Zeiss Contura (Carl Zeiss ltd., Rugby, UK) coordinate measuring machine (CMM). Measurements were taken using a 2 mm ruby stylus, recording 5 traces for each component at different heights from the cup rim, with an average of 14,684 points for each trace. Roundness values were automatically computed using the minimum zone circle method (difference between the radius of the most inside and outside points of the profile). The accuracy of the measurement technique is 2.2 + L/350 μm, where L is the measured dimension
Surface topography analysis
The surface topography of the internal cup surface was investigated using a contour GT-K 3D optical profilometer (Bruker, Coventry, United Kingdom), determining the roughness of the surface. A total of 10 measurements scans were taken, with a scan area of 437.1 μm × 582.7 μm, using a 20x objective lens and 0.55 multiplier. A Gaussian regression filter was applied to the raw data to obtain a measure of Ra roughness
Statistical analysis was performed using the statistical software package Prism (version 7.01, GraphPad, US). The data were assessed for normality using the D’Agostino-Pearson test. Non parametric Kruskal-Wallis and subsequent post hoc Dunn’s multiple comparison tests were used to determine significant differences among the three implants in pore size, lattice and solid thickness of the cup wall, roundness and roughness of the internal cup surface. One-way ANOVA and subsequent post hoc Tukey’s multiple comparison tests were used to determine significant differences among the three cups in porosity, strut thickness and overall thickness of the cup wall. The level of significance for all statistical analyses was p < 0.05.
The results are presented according to the location on the cup, starting with the analysis performed on the outer surface and continuing with the investigation of the internal surface of the three acetabular cup designs.
Summary of the median (interquartile range) measurements of the morphometric parameters for the three cups
72.3 (70.8 to 74.1)
76.4 (74.9 to 76.7)
63.0 (59.4 to 66.4)
Pore size, mm
0.841 (0.828 to 0.893)
1.004 (0.982 to 1.029)
0.521 (0.371 to 0.648)
Strut thickness, μm
472 (445 to 516)
648 (540 to 675)
421 (326 to 508)
Summary of the median (interquartile range) measurements of the cup wall thickness (dense, porous and overall) for the three cups
Solid thickness, mm
4.626 (4.242 to 4.805)
4.880 (4.774 to 6.095)
3.967 (3.839 to 4.033)
Lattice thickness, mm
1.347 (1.284 to 1.468)
1.468 (1.184 to 1.579)
1.038 (0.966 to 1.085)
Overall thickness, mm
5.973 (5.526 to 6.272)
6.424 (6.099 to 7.442)
4.955 (4.868 to 5.058)
Scanning Electron microscopy (SEM)
Surface topography analysis
This study is the first to compare different designs of 3D printed off-the-shelf acetabular cups from multiple manufacturers. We found a variability in the morphology of the outer surface due to the different 3D printing process (EBM vs LRM), with evidence of partially molten beads on all the cups. Differences in the lattice structures were also present, with a variability in the pore sizes and porosity of the cups, as well as in the thickness of the cup wall.
The analysis of the morphology of the outer surface of the cups from SEM images revealed the presence of partially molten beads on the struts of the porous structures. Smaller beads and higher beads density (beads/mm2) were found on the LRM-manufactured cup (Trident II) compared to the EBM-manufactured cups; this may be due to the smaller titanium powder beads used with the former compared to the latter .
These beads are a known by-product of the general 3D printing manufacturing process  however the presence of these in final-build acetabular components was noteworthy. The clinical impact of these partially molten beads needs to be understood, particularly if this may increase the release of titanium from the implants. Although titanium has superior biocompatibility, recent studies have revealed a series of adverse effects associated with this metal . To date, one study investigated the possibility of increased systemic titanium level in patients with a 3D printed off-the-shelf cup, finding no differences in comparison with a traditional titanium cup . However, the detection of titanium in blood samples is subjected to interferences from other chemical elements during the analysis; it has been recommended to use a high-resolution induced coupled plasma mass spectroscopy (ICP-MS) to obtain reliable outcomes, detecting titanium more accurately than other instruments . Further studies including 3D printed cups and using ICP-MS might help understand if the presence of the partially molten beads and the higher porosity of these acetabular components may be a concern for the patients. However, it cannot also be excluded that the presence of these beads may promote osseointegration, considering the rough surface that is created.
Three-dimensional printing allows design freedom that conventional manufacturing techniques cannot provide, as shown by the variability described for the three cups analysed using SEM and micro-CT.
The impact of pore size and shape on bone ingrowth and implant osseointegration is still a controversial subject and depends on the geometric and mechanical characteristics of the structure at the bone-implant interface. It has been suggested by in vitro and in vivo studies that pore sizes of both 100–400 μm and 500–1000 μm promote cell growth and proliferation [21, 22, 23]. Interestingly, the pores of the cups in this study were not circular in shape, which has been suggested to be more prone to occlusion . Although the specifications of the different pore shapes are proprietary, the methods to define the 3D porous structures are well established [9, 24]. As shown in this study, both regular (repeated unit cells) or irregular (stochastic) structures can be designed and manufactured. Human trabecular bone has an interconnected, open-porous structure with porosity of 50–90%, pore size in the order of 1 mm and trabecular (strut) thickness of hundreds of microns [25, 26]. The highly porous structure of the 3D printed cups showed values similar to those of human bone, with the ability of potentially reduce the risk of stress shielding due to the stiffness mismatch between implant and bone tissue [27, 28]. The hexagon-shaped porous structure, called Trabecular Titanium (TT), showed by the Delta TT cup has been previously characterized using cubic and cylindrical samples [17, 29, 30]. The values of porosity, pore size and strut thickness were comparable to our findings. Similarly, the values of porosity and pore size of the Trident II and Mpact cup were comparable to the specifications provided by the manufacturers [31, 32].
The design freedom of 3D printing enables thinner cup walls to be manufactured for a specific cup diameter. This means that a smaller cup can be chosen for specific head size, therefore sparing more bone stock. The three cups showed different dimensions both in thickness of the cup wall and depth of the porous structure.
The clinical impact of this variability in both the solid and lattice structures of the cups will be better understood from long-term clinical studies involving these 3D printed acetabular components. The Trident II and Mpact cups have recently been introduced on the market. The only study related to the Trident II cup compared the seating and the initial stability of this cup with the Trident I cup, which is conventionally manufactured, in an in vitro bench test using foam blocks . The Delta TT cup has been present since 2007 and clinical studies have reported satisfactory short to mid-term outcomes [14, 15, 16, 29].
While differences were found from the analysis of the outer surface of the cups, this was not true for the internal surface. The roundness and roughness of the internal cup surface was measured to identify potential differences in the dimensional and topographic properties of the cups, given by the 3D printing manufacturing method, that could influence the seating of the liner. As previously mentioned, the Trident II cup was produced using LRM, while the Delta TT and Mpact cups using EBM. Despite this difference, and the fact that even the EBM process can have distinct parameter settings between different manufacturers, the three cups showed similar roundness. This can be explained by the fact that 3D printing builds both the solid and lattice structure of the cups in a single stage layer-over-layer, but post-processing such as machining is performed to achieve the required tolerances to avoid dimensional mismatches with the liner and to guarantee its optimal accommodation. If the liner had an incorrect seating, a fracture (if ceramic) or an adverse effect on the fluid-film lubrication may occur, resulting in increased wear . The dimensional tolerance for this feature was not available, however, we found an overall mean roundness of 10.5 μm, suggesting near-perfect round shape of the internal surface of the cups.
No differences were found in the roughness of the internal cup surface. The post-processing of the cups managed to provide a Ra values of around 0.5 μm; it has been suggested that 3D printing parts for high end applications, such as orthopaedic implants, should have a surface roughness of less than 1 μm . An elevated surface roughness might influence mechanical wear and corrosion between the cup and the liner, because the reduced contact area between the two surfaces (internal of the cup, backside of the liner) might lead to increased localized stresses and more space for fluid ingress.
We acknowledge as limitations of this study the small cohort of acetabular cups under analysis and the non-uniformity in the cup size. The analysis of orthopaedic implants, especially if recently marketed, can be difficult. Further studies including more implants are needed to better understand the impact of the differences that we described.
This was the first study to compare different designs of 3D printed off-the-shelf acetabular cups from multiple manufacturers. We found a variability in the morphology of the outer surface of the cups and the properties of the porous structure. Although the existence of partially molten surface beads on 3D printed parts is a known by-product, their prevalence on these acetabular cups was interesting and the related clinical implications, if any, need to be investigated.
This comparison of different designs of 3D printed cups provides manufacturers and regulators, such as the Food and Drug Administration (FDA), the Medicine and Healthcare Products Regulatory Agency (MHRA) and the British Standards Institution (BSI) with evidence that may help to build robust investigation methods for this type of components and to monitor the implants that are already present on the market. Further laboratory studies, analysis of retrieved components and long-term clinical outcomes will help to prevent another metal-on-metal experience from happening.
The authors are grateful to Dr. Francesco Iacoviello for his valuable help during micro-CT data collection. This research study was funded by The Maurice Hatter Foundation, the RNOH Charity, the Rosetrees Trust and the Stoneygate Trust and supported by researchers at the National Institute for Health Research University College London Hospitals Biomedical Research Centre.
Conceptualization, L.D., H.H., J.H., A.D.L., P.S., A.H.; Design of the work, L.D., H.H., J.H., A.D.L., P.S., A.H.; Data acquisition and/or analysis, L.D., H.H., P.S.; Interpretation of data, L.D., H.H.; Manuscript writing and/or revision, L.D., H.H., J.H., A.D.L., P.S., A.H. All authors read and approved the final manuscript.
‘This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors’.
Ethics approval and consent to participate
Consent for publication
‘The authors declare that they have no competing interests’.
- 1.National Joint Registry for England, Wales NI and I of M (2018) 15th Annual Report. http://www.njrcentre.org.uk. .
- 2.Murr LE, Quinones SA, Gaytan SM, Lopez MI, Rodela A, Martinez EY, Hernandez DH, Martinez E, Medina F, Wicker RB. Microstructure and mechanical behavior of Ti-6Al-4V produced by rapid-layer manufacturing, for biomedical applications. J Mech Behav Biomed Mater. 2009;2:20–32. https://doi.org/10.1016/j.jmbbm.2008.05.004.CrossRefPubMedGoogle Scholar
- 3.Ryan G, Pandit A, Apatsidis DP. Fabrication methods of porous metals for use in orthopaedic applications. Biomaterials. 2006;27:2651–70. https://doi.org/10.1016/j.biomaterials.2005.12.002.CrossRefPubMedGoogle Scholar
- 4.Mumith A, Thomas M, Shah Z, Coathup M, Blunn G. Additive manufacturing. Bone Joint J. 2018;100-B:455–60. https://doi.org/10.1302/0301-620X.100B4.BJJ-2017-0662.R2.CrossRefPubMedGoogle Scholar
- 5.Hart AJ, Hart A, Panagiotopoulou V, Henckel J. Personalised orthopaedics – using 3D printing for tailor-made technical teaching , pre-operative planning and precise placement of implants. Orthop. Prod. News. 2017.Google Scholar
- 8.Murr LE, Gaytan SM, Martinez E, Medina F, Wicker RB. Next generation orthopaedic implants by additive manufacturing using electron beam melting. Int J Biomater. 2012;2012.Google Scholar
- 9.Wang X, Xu S, Zhou S, Xu W, Leary M, Choong P, Qian M, Brandt M, Xie YM. Topological design and additive manufacturing of porous metals for bone scaffolds and orthopaedic implants: a review. Biomaterials. 2016;83:127–41. https://doi.org/10.1016/j.biomaterials.2016.01.012.CrossRefPubMedGoogle Scholar
- 11.(2018) Mpact® 3D Metal™ Implants and Augments 3D Metal™ (K171966)Google Scholar
- 12.(2016) Trident® II Tritanium® Acetabular Shell (K161569)Google Scholar
- 13.(2012) Delta TT Acetabular System (K112898)Google Scholar
- 22.Bobyn JD, Pilliar RM, Cameron HU, Weatherly GC. The optimum pore size for the fixation of porous-surfaced metal implants by the ingrowth of bone. Clin Orthop Relat res 263–270. 1980. https://doi.org/10.1097/00003086-198007000-00045.
- 23.Van Bael S, Chai YC, Truscello S, Moesen M, Kerckhofs G, Van Oosterwyck H, Kruth JP, Schrooten J. The effect of pore geometry on the in vitro biological behavior of human periosteum-derived cells seeded on selective laser-melted Ti6Al4V bone scaffolds. Acta Biomater. 2012;8:2824–34. https://doi.org/10.1016/j.actbio.2012.04.001.CrossRefPubMedGoogle Scholar
- 24.Leary M (2018) Design of titanium implants for additive manufacturing. In: Titanium in Medical and Dental Applications. Elsevier Inc., pp 203–224.Google Scholar
- 25.Keaveny TM, Morgan EF, Niebur GL, Yeh OC (2001) Biomechanics of Trabecular Bone. Annu Rev Biomed Eng 3:307–33 . doi: org/ https://doi.org/10.1146/annurev.bioeng.3.1.307
- 26.Koschwanez HE, Reichert WM (2013) Textured and porous materials. In: Biomaterials science: An introduction to materials: Third Edition, Third Edit. Elsevier, pp 321–331.Google Scholar
- 27.Karageorgiou V, Kaplan D. Porosity of 3D biomaterial scaffolds and osteogenesis. Biomaterials. 2005;26:5474–91. https://doi.org/10.1016/j.biomaterials.2005.02.002.CrossRefPubMedGoogle Scholar
- 28.Murr LE, Gaytan SM, Medina F, Lopez H, Martinez E, MacHado BI, Hernandez DH, Martinez L, Lopez MI, Wicker RB, Bracke J. Next-generation biomedical implants using additive manufacturing of complex cellular and functional mesh arrays. Philos Trans R Soc A Math Phys Eng Sci. 2010;368:1999–2032. https://doi.org/10.1098/rsta.2010.0010.CrossRefGoogle Scholar
- 30.Regis M, Marin E, Fusi S, Pressacco M, Fedrizzi L. Preparation and characterization of newly developed trabecular structures in titanium alloy to optimize Osteointegration. Int J Medical, Heal Pharm Biomed Eng. 2014;8:279–84.Google Scholar
- 31.www.stryker.com (accessed 01/04/2019).
- 33.Davignon R, Rajaravivarma R, Yanoso-Scholl L, Gopalakrishnan A, Abitante P (2017) Comparison of seating and initial stability of an additive manufactured porous Acetabular Shell to a clinically successful porous Acetabular Shell. In: 30th annual congress of the International Society for Technology in Arthroplasty. Pp 146–148.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.