8 Biomechanical testing of different plates
Adequate stable fixation is mandatory for safe healing of both additive and subtractive osteotomies around the knee, in order to minimize the risk of nonunion and loss of correction. It has been shown that for open-wedge high-tibial osteotomy (HTO), short spacer plates (Puddu plates) have a large failure rate and a high risk of implant or screw breakage and loss of correction.
Angular stable implants (internal fixators) have replaced conventional plate systems in operative fracture management to a large extent. The highly improved stability of these plates and the possibility of to use them as a “biologic” osteosynthesis with only a minimal approach (MIPO) has made them superior to the non-angular stable plates in many applications.
Due to the long lever arms of the lower extremity and the risk of nonunion, osteotomies around the knee always have been problematic [1–3]. The established system of angular stable osteosynthesis was therefore applied to the concept of osteotomies around the knee, and the plates were specially adapted and precontoured to fit the lateral and medial anatomy of the proximal tibia and distal femur.
This chapter describes a set of experiments that were carried out to test the primary fixation strength of the implants of the TomoFix group for the medial proximal tibia.
In a first series the original TomoFix plate (TomoFix standard implant) was comparatively tested against three other implants (spacer plates), which are available on the market for fixation of HTO . In two following series a modified TomoFix plate (TomoFix next generation, NG), which was developed and designed by the Knee Expert Group (KNEG) of the AO and by Synthes, and a special new small plate of TomoFix (TomoFix small version), designed for the Asian population and small patients, were tested. The results were compared to existing data for the implants.
2.1 Mechanical testing system ( Fig 8-1 )
Fifteen (n = 15) third generation large composite tibiae were used in this study (saw bones Europe AB, Malmo, Sweden). The tibiae have been reported to reproduce the structural mechanical properties of healthy young adult tibiae [5, 6]. They were embedded in a two component polyurethane casting resin (UREOL FC 53, Vantico GmbH, Wehr, Germany) at their proximal and distal ends, and mounted in a specially designed fixture which was attached to a materials testing machine (MiniBionx 858, MTS Systems Corporation, Eden Prairie, MN, USA). The fixture was constructed to allow standardized axial loading at a 62% lateral offset from the center of the joint as recommended for clinical use  ( Fig 8-2 ). The tibial plateau was embedded in a shallow mold so that force could be applied and different osteosynthesis plates attached without interference. Load was applied to the tibial plateau by means of a ball joint which allowed complete rotational freedom of motion. The distal end of the tibia was potted in a cylinder attached to a universal joint which allowed frontal and sagittal plane rotation but constrained axial rotation.
2.2 Implants tested in first series ( Fig 8-3 )
Four different fixation devices were compared ( Table 8-1, Fig 8-3 ): a short spacer plate without locking bolts (open-wedge osteotomy plate, OWO) (n = 4), a short spacer plate with multidirectionally insertable locking bolts (multidirectional angular stable osteotomy plate, MSO) (n = 5), a long spacer plate with multidirectionally insertable locking bolts (MSOnew) (n = 2), and the TomoFix standard fixator (n = 4).
2.3 Osteotomy ( Fig 8-4 )
All osteotomies were performed using exactly the same technique and by the senior author according to the standardized technique recommended by the KNEG of the AO (described in chapter 9 “High-tibial open-wedge valgization osteotomy with plate fixator”) . This comprised a biplanar osteotomy with a horizontal cut of the posterior 2/3 of the tibia. This cut was incomplete leaving 10 mm of lateral bone intact (lateral hinge), which served as the pivot-point during the opening of the osteotomy. The second and complete cut was carried out in the anterior third of the tibia in the frontal plane, ascending 110° behind the tibial tuberosity. The surface areas of this frontal-plane cut maintained direct contact during the opening of the osteotomy ensuring correct axial and rotational alignment of the osteotomized bone. The opening was carried out slowly and carefully over several minutes using the 3-chisel technique in order to avoid microfractures in the lateral cortex during the process of bending around the hinge. A standardized opening of 10 mm was created in all tested saw bones.
After opening the osteotomies, all plates were mounted according to the manufacturers’ guidelines using the original operative instruments. In the three types of plates with spacer blocks (OWO, MSO, MSOnew), a direct contact of the spacer blocks with the adjacent surface of the osteotomy gap was assured, so that loading was transmitted through the bearing surface of the spacer. The distal cortical screws of OWO, MSO, and MSOnew were inserted bicortically, and the proximal cancellous screws were inserted at an angle of 90° to the plate. The screws of the MSO and MSOnew plates were locked into place using the instruments designed by the manufacturer for that purpose. The TomoFix standard plate was attached to the bones using the original small gap spacers as well as the drilling guides to ensure appropriate locking of all bolts within the plate. The three most proximal screws, as well as the screw distal to the osteotomy, were inserted bicortically. The three most distal screws were inserted monocortically.
2.5 Displacement at osteotomy gap ( Fig 8-5 )
After attachment of the plates, the tibiae were embedded in the mounting fixtures and inductive linear displacement transducers (SM277, Schreiber Messtechnik, GmbH, Oberhaching, Germany) were attached at the medial and lateral sides of the osteotomy gap using bone cement in a reproducible manner with a small alignment fixture. The displacement transducers allowed measurement of the motion between the proximal and distal segment of the osteotomy with a spatial resolution of 0.01 mm.
2.6 Testing protocol ( Fig 8-7 )
The specimens were tested according to two protocols (see Table 8-1 ); the first consisted of two nondestructive low-load cyclical preloading cycles followed by a displacement-controlled single-load to failure, the second of a load-controlled cyclical staircase fatigue loading protocol to failure. The prefailure loadings at low load were intended to simulate a patient either partially or fully weight bearing immediately after surgery. Two sinusoidal loadings were thus applied for 60 cycles at 0.25 Hz and a loading level of 150N and 800N. The specimens were subsequently destructively loaded to failure at 1 mm/s and the displacement across the osteosynthesis gap during failure was measured. The first point of failure (PF1) and its corresponding loading were defined as the point at which the first reduction in loading occurred (first peak, Fig 8-7 ), the second point at which the maximum loading occurred (PF2, Fig 8-7 ). For the specimens loaded to fatigue failure, a staircase dynamic loading protocol was defined based on the average ultimate load of the specimens subjected to a single-load to failure (~1600N). The initial loading level was set to 50% of that load (800N), and loading was increased in 10% steps after the successful completion of each loading step of 20,000 cycles (loading levels: 800N (LL1), 960N (LL2), 1120N (LL3), 1280N (LL4), etc; loading frequency = 2Hz). Loading was terminated after actuator displacements of more than 2 mm were observed during one loading cycle, which was defined as failure of the construct, and the total number of cycles as well as the loading step were recorded.
2.7 Residual stability after failure
The stiffness of the different bone-implant constructs after failure was tested comparatively in order to determine the residual stability after breakage of the lateral cortex. The proximal and distal segment of the osteotomy with the plates still in place, but with failed lateral cortex, were manually twisted and bent, and the resulting displacements at the lateral cortex were measured with a ruler.
3 Results of first series with TomoFix standard
3.1 Low-load cyclical tests
Axial cyclical loading with 150N and 800N was tolerated by all implants without failure. No visible damage to saw bone, implant, or the saw-bone implant interface was found. Sinusoidal steady state displacements were measured at both the medial and lateral osteotomy gap secondary to the force which was cyclically applied by the testing machine at 0.25 Hz for 60 cycles. Loading with 800 N created a displacement with a maximum amplitude of 0.13 mm medially and 0.44 mm laterally. These displacements were constant over all 60 cycles in all tested bones and in all implants.
3.2 Destructive single-load to failure tests
All tested bone-implant constructs in the single-load to failure tests (n = 6) failed at the lateral cortex of the saw bone with an identical failure mode (see Fig 8-5 ). First a fissure in the lateral cortex was observed creating a sudden drop in the applied loading force measured (see Fig 8-6 ). Another major drop-off of the loading force was typically observed as total failure of the lateral cortex with manifest breaking of the entire lateral bone hinge occurred (see Fig 8-6 ). Apart from MSO, where additional fissures in the bone around the distal and anterior screw were noted at the time of PF2 (see Fig 8-7 ), no fractures or breakages in any other area of the saw bone were noted in any of the tested bone-implant constructs. In OWO and MSO, an additional flexion displacement of the proximal segment of the osteotomy in the sagittal plane was observed at PF2 ( Fig 8-8 ), which was not observed in TomoFix standard.