Abstract
Objectives:
Tibial spine fractures (TSFs) are common in the pediatric population. While screw and suture fixation are used for surgical repair, most studies in porcine or adult human bone suggest suture fixation is superior. However, a recently published report by this study group demonstrated that two-screw and two-suture repair, where sutures were tied over a metaphyseal bone bridge, were biomechanically comparable in human pediatric cadaveric knees. The key failure mechanism of the suture group involved ‘cheese wiring’ through the metaphyseal bone bridge. Therefore, this study hypothesized that TSF fixation with sutures attached to anchors placed in stronger cortical bone would produce biomechanically superior repair. It evaluated the properties of these constructs in human pediatric knees and compared them to prior results obtained from two-screw and two-suture constructs.
Methods:
In the previous study, pediatric knee specimens were randomly assigned to two-screw or two-suture fixations. Circumferential marks were made around the ACL insertion (Figure 1A). An osteotome induced a standardized Meyers-Mckeever Type III TSF (Figure 2A). For knees assigned to screw fixation, each fracture was reduced and held in place with two 4.0 mm cannulated screws and washers (Figure 1B). For knees assigned to suture fixation, two No. 2 high-tensile-strength sutures were passed through the fracture fragment and the base of the ACL, one through the anterior 1/3, the other through the posterior 1/3 (Figure 2B). The sutures were passed through medial and lateral bony tunnels, each 1cm away from the midline and two physis lengths distal to the tibial plateau. The fracture was reduced by pulling tension on the sutures, which were then secured over a 1cm tibia bony bridge (Figure 2C). Each specimen was potted in fiberglass resin and then mounted for biomechanical testing on a servohydraulic load frame (Figure 1D, 2D) at approximately 30° of flexion to simulate typical ACL loading conditions. Each specimen was then subjected to cyclic preconditioning, which consisted of 20 cycles of loading between 5 and 25N at a rate of 60 cycles per minute. Next, a cyclic loading protocol was applied to each specimen. This included 500 cycles between 5 and 75N at a crosshead speed of 100mm per minute, sampled at 20Hz. Upon the completion of cyclic loading, samples were allowed to recover for thirty minutes.
Finally, a load-to-failure protocol was conducted at a rate of 0.5 mm per second. The primary outcome was the ultimate failure load in newtons (N). The mode of failure was macroscopically documented. In the current study, six new pediatric knees were selected to best match the age and laterality of the previous screw and suture groups. These specimens were placed through the same dissection, fracture, biomechanical testing, and data collection protocols used in the prior study. However, the present study assessed a different fixation: suture plus suture anchors. Suture plus suture anchor specimens were repaired in the same way as suture specimens, with the following exception: instead of tying sutures across a metaphyseal bony bridge following their exit of the medial and lateral bony tunnels, the sutures were secured with knots to two 2.8 mm suture anchors, also placed 1 cm away from the midline medially and laterally, at a distance of three physis lengths from the tibial plateau (Figure 3A, 3B, and 3C). Independent samples t-tests compared ultimate failure loads between pairs of repair methods, while a one-way analysis of variance test was used to compare ultimate failure loads across all repair methods.
Results:
A total of eighteen (Range: 5 years – 11 years) pediatric cadaveric knees were tested: twelve from the prior published study, and six in the suture plus suture anchor group. The screw and suture repair groups had identical mean (8.30 years) and median (8.50 years) ages, while the suture plus suture anchor group had a mean and median age of 9.33 and 9.00 years, respectively. All groups had an identical number of samples of each laterality. The ultimate failure load significantly differed across fixation methods (p = 0.008). This relationship was primarily driven by higher ultimate failure loads in the suture plus suture anchor group (Mean: 224.00N, SD: 48.37N) when compared to the screw (Mean: 143.52N, SD: 41.97N) (p=0.01) and suture (Mean: 135.35N, SD: 47.94N) (p=0.009) groups. Ultimate failure load did not significantly differ between screw and suture fixations (p = 0.7597). One screw and one suture construct did not survive the cyclic loading protocol. All suture plus suture anchor constructs survived. Every screw construct failed at the level of the intact tibial spongy bone; no failure occurred through the ACL-fracture fragment complex. Four suture constructs failed by cheese wire through the cortical bony bridge, while two failed by cheese wiring, thus creating a new fracture, in the ACL-tibial spine avulsion fracture complex. Each suture plus suture anchor construct failed because of a fracture of the ACL-tibial spine avulsion fragment complex (Figure 3D).
Conclusions:
Screw and suture fixation of type III Meyers-Mckeever TSFs have statistically comparable ultimate failure loads in human pediatric cadaveric bone. These loads of failure are relatively low compared to those obtained in adult cadaveric and porcine bone. This study demonstrates that sutures plus suture anchors placed in the meta-diaphyseal cortex provide a significantly stronger fixation. Although this study is limited by its inability to assess rotational biomechanics and slight age differences between fixation groups, these results suggest that suture plus suture anchors provide the strongest repair for these injuries in human pediatric specimens.
