Page 507 - IJB-10-1
P. 507

International Journal of Bioprinting                          Macro and micro structure of a 3D-printed implant




            of our assembly bone plate should meet the requirements   secured with a specific square clamp for the distal part and
            for implant manufacture.                           a circular fixture for the proximal part on a material testing
                                                               machine  (Instron  E10000,  Instron,  Canton,  MA,  USA).
            2.5. Biomechanical fatigue test                    Two medial and lateral extended plates attached onto the
            The biomechanical fatigue test was conducted according   proximal circular fixture were also prepared to measure the
            to the literature to compare the mechanical performance   corresponding displacement on both sides using miniature
            using the OWS and the commercial HTO TP fixation   linear potentiometers (Novotechnik T-0025, U.S. Inc.,
            system by gradually increasing the load to simulate   Germany) (Figure 6c).
            complete weight-bearing during the gait cycle. 30,31  The test
            samples were prepared using tibia sawbone material (Tibia,   The experimental fatigue test involved a multi-stage
            4th Gen., Composite, 17 PCF Solid Foam Core, Medium)   cumulative loading procedure with increments of 160 N
            to produce osteotomy models that were cut using a metal   and was subjected to 20,000 cycles of dynamic loading at
            saw machine to simulate a 12° corrective defect. The   a frequency of 5 Hz at each stage. The loading started at
            3D-printed OWS and the commercial TP system (A Plus   800 N and gradually increased up to 5440 N, completing
            Biotechnology Co., Ltd., New Taipei City, Taiwan Region)   30 load stages or until the test samples failed. The medial
            were subsequently fixed separately at the osteotomy site for   and lateral linear potentiometer displacement values were
                                                               recorded during the experiment. Two failure criteria were
            each group consisting of three sets (Figure 6a and b).
                                                               established: (i) a medial-lateral displacement difference on
               Regarding the fatigue experimental setup, the distal   the tibial plateau exceeding 2 mm, and (ii) the occurrence
            and proximal tibia were embedded in epoxy resin and   of a visible collapse fracture on the lateral cortical bone,















































            Figure 6. (a,b) Optimal wedge-shaped spacer (OWS) and T-shaped plate (TP) systems on sawbones; and (c) illustration of the biomechanical fatigue test
            setup.


            Volume 10 Issue 1 (2024)                       499                          https://doi.org/10.36922/ijb.1584
   502   503   504   505   506   507   508   509   510   511   512