1 Introduction

Addressing a posterior column (PC) acetabular fracture from an anterior based approach, specifically when treating the associated both column and anterior column posterior hemitransverse patterns, presents a unique challenge for reduction and definitive fixation. Achieving an anatomic reduction requires compression across the fracture while also controlling fracture rotation through fracture keys on the quadrilateral surface. While there is no established gold standard, multiple reduction and fixation strategies have been developed with the common goal of anatomic restoration and stable fixation.^1,2 Often, the posterior column can be reduced from an anterior approach using indirect reduction techniques, but fixation strategy requires more creativity than through a posterior approach where direct access to the plating surface is readily available. Despite these challenges, the posterior column should not be overlooked as inadequate reduction and fixation strength of acetabular fractures involving the posterior column may result in medial subluxation, malunion, dislocation, arthrosis, and pain.^3,4

A suprapectineal pelvic reconstruction plate and screw construct (P&S) remains the most traditional method for posterior column stabilization from the anterior approaches as the two, 3.5 mm interfragmentary screws are thought to provide rotation control. In addition, using the smaller core diameter screws allows for ease of placement perpendicular to the primary fracture line. Anecdotally, an alternative fixation technique using a 6.5-mm partially threaded cannulated posterior column screw (PCS) placed from the pelvic brim to the ischial tuberosity provides outstanding interfragmentary compression without compromising rotational control. Furthermore, because the 6.5 mm screw volumetrically fills the posterior column, there may be less fracture site motion with early mobility.

The present study aims to establish whether the P&S construct or PCS construct is biomechanically superior for posterior column fixation in acetabulum fractures approached from the anterior side. We hypothesize that the PCS construct will be more effective than the P&S construct in minimizing relative motion of fracture fragments. We suspect that filling the osseous corridor of the posterior column with a large diameter (6.5-mm) screw may have a biomechanical strength advantage and reduce interfragmentary motion in fixation of the posterior column.

2 Methods

2.2

2.2 Specimen preparation

In order to create reproducible posterior column fracture lines in each hemipelvis model, as well as comparable screw trajectories for each construct, the authors fabricated clay molds (Sculpey Air Dry, Elk Grove Village, IL). For the screw trajectory molds (one for each construct), this was accomplished by first drilling appropriate screw paths in one hemipelvis model, and then molding the clay around the 2.5 mm pelvic drill bit(s) or 2.8 mm guidewire, respectively, and around nearby osseous anatomy as reference points. These molds were allowed to dry completely before being removed. The appropriate mold was placed on each subsequent hemipelvis, and the drill bits or guidewires then placed within the hollow cylindrical paths in the mold to recreate the same precise screw trajectories (see Fig. 1).

Close

Fig. 1 – Screw Path Trajectories Clay molds were utilized to create reproducible screw path trajectories. The black arrow represents the posterior column screw trajectory, while the white arrow represents the trajectories for the plate and screw construct.

Export to PPT

For the fracture line mold, a posterior column fracture line was drawn on one hemipelvis model with indelible ink. A clay mold was then fitted within the acetabulum, filling only the portion of the acetabulum superior to the drawn fracture line, and allowed to dry completely. This mold was then utilized to draw identical fracture lines through the posterior column in each Sawbone; the line was then extended superiorly toward the greater sciatic notch and inferiorly through the inferior pubic ramus.

Prior to making the fractures, each hemipelvis was potted in a 2-part epoxy resin (Bondo, 3M, Maplewood, MN), such that the sacroiliac joint was encased. This allowed each hemipelvis to be held firmly in a multiaxial vice grip while a precision thin saw blade (9.0 × 0.38 × 25 mm) was used to create the posterior column fractures (see Fig. 2). The fractures were then clamped in anatomic position using a standard surgical clamp, and implants for the appropriate construct were placed through the pre-drilled paths (see Fig. 3).

Close

Fig. 2 – Fracture Line Morphology After using a clay mold to draw identical fracture lines through the posterior column of the acetabulum, the drawn fracture lines were then extended superiorly toward the greater sciatic notch, and inferiorly through the inferior pubic ramus.

Export to PPT

Close

Fig. 3 – Implant Placement Each specimen was mounted in a rectangular block of epoxy resin at the sacroiliac joint prior to creation of the fracture lines. After the fracture was clamped, implants were placed in each hemipelvis.

Export to PPT

2.4

2.4 Loading

Testing was conducted on an Instron E3000 Linear-Torsion All-Electric Dynamic Test Instrument (Instron E3000, Norwood, MA). The testing setup is depicted in Fig. 4. The pelvis models were cyclically loaded for 100 cycles (0.5 cycles/second) at both 400 and 800N, first in a sit-to-stand position that is expected during recovery, and subsequently in a squat-to-stand position to test overload conditions ¹². Dynamic force-controlled loads were applied in a sinusoidal pattern ranging from 20 N to the desired load, at a rate of 0.5 cycles/second. The load magnitudes were based on Hempen et al.,¹³ in which 400 N simulated light use during patient recovery, and 800 N was an “overload”. To interface with the specimen acetabulum, a 2″ Acetal Resin ball apparatus was mounted to the Instron actuator. Two linear bearings (model HSR 20C, THK, Schaumburg, IL) allowed free horizontal movement of the ball.

Close

Fig. 4 – Testing Setup The sawbone specimens were held in a multi-axial vice by the epoxy resin mount. Motion tracking sensors were placed on either side of the fracture site. A 2″ Acetal Resin ball apparatus was mounted to the Instron actuator.

Export to PPT

3 Results

The PCS construct resulted in significantly less relative motion at the fracture site than the P&S construct under sit-to-stand loading (0.06 ± 0.02 mm vs 0.1 ± 0.02 mm at 400N, p = 0.02; 0.13 ± 0.03 mm vs 0.19 ± 0.04 mm at 800N, p = 0.03). The PCS construct also demonstrated significantly less relative fracture site motion under squat-to-stand loading (0.22 ± 0.13 mm vs 1.9 ± 0.5 mm at 400N, p = < 0.001; 1.48 ± 0.44 mm vs 4.77 ± 0.3 mm at 800N, p = < 0.001). These results are detailed in Table 2 and depicted in Fig. 5 (see Fig. 5 and Table 2).

Close

Fig. 5 – Results: Relative Motion at Fracture The PCS construct resulted in significantly less relative movement at the fracture site during both sit-to-stand and squat-to-stand testing.

Export to PPT

All 12 specimens withstood sit-to-stand testing at both 400N and 800N, as well as squat-to-stand testing at 400N (see Fig. 4). During squat-to-stand loading at 800 N, 6 of 12 repairs (50 %) reached a cutoff point of 10 mm of relative movement, at which point the repairs were considered to have failed. These specimens consisted of 2 PCS repairs and 4 P&S repairs (see Table 3). Tests were ended when significant movement in excess of 10 mm could be observed, and the exact point at which 10 mm of movement first occurred was determined analytically in Matlab.

4 Discussion

In the present study a single 6.5-mm partially threaded cannulated screw demonstrated less relative fracture motion in the sit-to-stand and squat-to-stand test conditions as compared to a 7-hole pelvic reconstruction plate and four 3.5-mm cortical screws for fixing posterior column acetabulum fractures from an anterior based approach. The PCS construct withstood greater cyclic loading in both sit-to-stand and squat-to-stand positions as well as a less likelihood of undergoing catastrophic failure. Theoretical benefits of the single 6.5-mm screw are: 1) the screw traverses the entirety of the posterior column corridor length from the brim to the ischial tuberosity and, 2) it volumetrically fills the column by contacting multiple points in the bony corridor, in similar fashion to an intramedullary nail.

There are very few prior studies that specifically evaluate posterior column fixation strength from an anterior based approach. Schopfer et al. investigated posterior column fixation strategies from a posterior approach including: 1) one 3.5-mm reconstruction plate, 2) two 3.5-mm reconstruction plates, and 3) a 4.5-mm lag screw with a single plate. The sole significant difference they identified between the three constructs was a smaller interfragmentary compliance as a result of fixation with a lag screw and neutralization plate (option 3) at 60 degrees of flexion.¹⁴ More recently, Su et al. investigated posterior column fixation strategies with posteriorly based fixation constructs: 1) reconstruction plate, 2) reconstruction plate with a PC lag screw, 3) two reconstruction plates, and 4) W-shaped angular plate. They demonstrated that the W-shaped angular plate was the biomechanically stiffest construct, and partially restored joint loading parameters toward an intact state.¹⁵

Wang et al. compared 5 separate anteriorly and posteriorly based constructs utilizing a computerized finite element analysis (FEA) model: 1) reconstruction plate, 2) PC screw, 3) infra-acetabular screw, 4) infra-acetabular screw and reconstruction plate, and 5) one infra-acetabular screw and one PC screw. Based on their model, fixation with both a PC screw and an infra-acetabular screw (option 4) allowed for column filling fixation that provided superior mechanical support.¹⁶ The present study expands on this FEA by extrapolating into a tangible model and supports Wang et al. in that the posterior column lag screw was biomechanically superior to a plate and screw construct in all parameters.

Although the 6.5-mm screw fit safely within the posterior column in our sawbones model, there remains concern that with anatomic variation in vivo the bony corridor would accommodate such a large screw. To this aim, Shahulhameed et al. carried out a cadaveric analysis to evaluate the average dimensions of both the anterior and posterior columns of the acetabulum and to determine which diameter screws could be safely passed utilizing percutaneous techniques. With regard to the posterior column, they found a mean thickness of 21.3-mm, and postulated that 7.3-mm screws could be safely placed within most posterior columns.¹⁷ As our study utilizes a smaller diameter 6.5-mm screw, it can be inferred that this screw would also be of an appropriate size for most adult patients. Anecdotally, the senior author (MM) has similarly found that the corridor of the posterior column, from the pelvic brim to the ischial tuberosity, accommodates a 6.5-mm screw in most adult patients.

While the present study utilized cannulated screws placed only in an antegrade fashion to best mimic placement through an anterior based surgical approach, 6.5 mm cannulated screws may also be inserted in a retrograde fashion via percutaneous methods. Percutaneous fixation of the posterior column has been thought to be advantageous in certain circumstances, such as for non-displaced or minimally-displaced fractures, and/or in particular patient populations such as the obese and elderly.¹⁸ Percutaneous fixation methods are also associated with less soft tissue disruption, shorter length of surgery, and decreased blood loss.¹⁹ Preliminary results in the literature have shown impressive results after percutaneous posterior column fixation. Levin et al. demonstrated a union rate of 100 % with no instances of implant violation of the hip joint or sciatic notch after percutaneous posterior column fixation in a prone position.²⁰ In minimally displaced posterior column fractures or in associated fracture types that are secondarily reduced with reduction of the anterior column, it is advantageous to be able to rigidly fixate the posterior column with a percutaneously-placed column filling screw – without the increased biological insult of a large lateral window incision for application of a suprapectineal plate.

There is also an important cost difference between the two tested constructs. The 6.5-mm screw used in this study costs approximately $231 USD, in comparison to $421 USD for the combined cost of a reconstruction plate and four 3.5-mm screws. While there are certainly small differences in cost among implant companies, this cost differential is likely similar, and has important implications for overall cost-of-care for patients.

There are a few key limitations of this study. First, this is a biomechanical study using composite bony specimens. Although we used a standardized testing protocol that we believe best simulates reality, it does not fully account for muscular and other physiological in vivo forces. Second, our study has a small sample size of 12 total hemipelvi. Although additional specimens certainly would have strengthened our findings, we easily reached statistical significance in all testing parameters. Lastly, although extensive measures were taken to reproduce comparable fracture lines and screw trajectories among the specimens, it is still possible that there was some minor variety in implant placement. However, a small degree of variety amongst specimens is certainly analogous to true clinical scenarios, in which there are small differences between individual fracture morphology, patient anatomy and surgical technique.

In conclusion, fixation of a posterior column fracture with a single 6.5-mm partially threaded cannulated screw is a robust, biomechanically stable, biologically advantageous, and cost-effective method of fixation. The PCS fixation technique withstands comparable physiologic forces to a standard P&S construct which may prevent unwanted fracture site motion in the early postoperative period. This biomechanical analysis demonstrates equipoise for prospective analyses exploring the potential safety of earlier advancement of weight bearing for acetabulum fractures involving the posterior column when the PC fracture line is repaired with a PCS construct.

Ethical approval statement

Neither IRB approval nor patient consent was required for this project, as it did not involve patients or patient information. Solely synthetic hemipelvis models were used for our biomechanics study. The ethical principles outlined in the Declaration of Helsinki (2013) are thus not applicable either.

Funding

This work was supported by an AO Trauma North America Fellows Grant (Grant # 171408-FEL2021).

Guardian/patient consent statement

N/A; no patients were investigated during this study.

CRediT authorship contribution statement

Helyn G. Fraser: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Visualization, Writing – original draft, Writing – review & editing. M. Wesley Honeycutt: Conceptualization, Funding acquisition, Methodology, Project administration, Resources. Daniel Thompson: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft. Nina Suh: Data curation, Formal analysis, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – original draft, Writing – review & editing. Huai Ming Phen: Conceptualization, Investigation, Methodology, Project administration, Resources, Writing – review & editing. William S. Godfrey: Conceptualization, Investigation, Methodology, Project administration, Resources, Writing – review & editing. Adam R. Boissoneault: Conceptualization, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing. Michael A. Maceroli: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.