1 Introduction

Tibial pilon fractures represent a complex array of injuries, frequently associated with distal fibular fractures. Representing a spectrum of high and low energy injuries, anatomic restoration of the lateral column for any associated fibular fracture is a fundamental principle of surgical treatment as initially described by Ruedi and Allgower.¹ Fixation of the fibula in the setting of poor soft tissue envelope, however, is a challenging problem. In cases of tibial fracture with comminution, the soft tissues can be severely damaged or at risk, thereby precluding formal open reduction and plating. As these injuries frequently represent an associated circumferential injury, open fixation according to the principles outlined by Ruedi and Allgower has been associated with high complication and wound infection rates. For these reasons, alternative options, including the use of staged surgical fixation and/or various percutaneous techniques, which decrease the number of skin incisions, have been proposed.²

The options with percutaneous fixation of the fibula include locked intramedullary nailing, flexible nailing, smooth Steinman pinning and intramedullary screw fixation. These percutaneous techniques have been compared to traditional plate fixation and have demonstrated similar functional outcomes with the benefit of significantly decreased wound complications.^3,4 Theoretically, the bite of the screw into the endosteal bone provides a more stable fixation, while smooth implants may provide fixation with some interfragmentary motion, supporting secondary bone healing. In the current study, we compare the biomechanical quality of threaded versus smooth intramedullary fixation of the fibula. We aim to determine which percutaneous fixation modality provides more stability to a transverse fibula fracture associated with 43A and 43C fractures according to the AO Foundation/Orthopaedic Trauma Association (AO/OTA) fracture classification.⁵ The null hypothesis is that there would be no significant difference in stability between threaded screw and smooth pin fixation at the fibular fracture site.

2 Methods

Nine pairs of matched cadaveric lower extremities below the knee were obtained after institutional approval and stripped of all soft tissue attachments other than the ankle capsular and syndesmotic ligaments above the malleoli. Using a thin-bladed sagittal saw, a simple transverse fibula fracture was created in the suprasyndesmotic space of each fibula, 7 cm above the tip of the lateral malleolus. With randomization of the left or right side by means of a function generator (Matlab), one specimen from each pair of fractured fibulas was repaired with an intramedullary threaded screw and the contralateral side was repaired with a smooth flexible nail. Both the screw and nails were placed flush within the tip of the distal fibula. For intramedullary fixation with screw fixation, Synthes 4.5 mm stainless steel, cortical screws were used. These screws were fully-threaded, non-self-drilling and 150 mm in length which were the longest available to us. Cortical screws were chosen to maximize diaphyseal fit. Screw diameter and length was determined by fibular diameter measurement on preoperative plain radiograph or computed tomographic (CT) scan. For intramedullary fixation using smooth nails, Synthes 3.0 mm flexible titanium nails that were 120 mm in length were used (Fig. 1).

Close

Fig. 1 Schematic of the fully threaded screw (example from an operative pilon fracture case) vs. flexible titanium nail fixation used for intramedullary fixation of the distal fibula fracture.

Export to PPT

Each repaired fibula construct was placed into a custom jig with polyester resin anchoring the foot and mounted in a servohydraulic load frame (MTS Bionix 858) (Fig. 2). Each specimen was tested in both 5 N-m cyclic external loading combined with 700 N constant axial loading and in 700 N cyclic axial loading. Three-millimeter diameter black spherical markers were attached on both sides of the fracture site and movement at the fracture site was assessed with video dimensional analysis to track both vertical and shear displacement during the cyclic loading. The test protocol applied 2000 cycles of 0–5 N-m axial torque under constant 700 N vertical load and then 500 cycles of vertical loading between 100 N and 700 N. The displacement at the fracture site in both planes was measured at the start and end of each loading cycle.

Close

Fig. 2 Testing specimen loaded in polyester resin mold in mounted tester with VDA (video dimensional analysis) reference points.

Export to PPT

The displacement in shear (parallel to the fracture) and axial (vertical or perpendicular to the fracture) directions were both measured for statistical analysis. These values were assessed at both initiation and the end of the loading protocols (Fig. 3). Analysis of variance (ANOVA) testing was performed once for axial loading and once for external rotation with axial loading. Each ANOVA had two factors: time (before or after load cycle) and repair type (threaded screw or flexible nail).

Close

Fig. 3a Shear displacement (mm) of pre and post threaded screw fixation and pre and post flexible titanium nail fixation used for intramedullary fixation of the distal fibula fracture.

Export to PPT

Close

Fig. 3b Axial (vertical) displacement (mm) of pre and post threaded screw fixation and pre and post flexible titanium nail fixation used for intramedullary fixation of the distal fibula fracture.

Export to PPT

3 Results

One pair of specimens was lost to catastrophic failure with instrumentation and mounting during insertion of a threaded screw, leaving 8 pairs for statistical analysis (n = 8). In vertical loading alone, both screw and smooth pin fixation demonstrated similar outcomes, with no significant differences according to the ANOVA for time (p = 0.719), repair type (p = 0.844) or interaction of time and repair type (p = 0.876) (Fig. 4a). However, with external rotation stress at the ankle, screw fixation demonstrated significantly less shear displacement at the fracture site than flexible nail fixation (p = 0.018). In our additional assessments of shear displacement, comparison of preoperative and postoperative fixation demonstrated no significant difference (p = 0.853) and the interaction of repair type and time was also not significant (p = 0.419). The vertical displacement in axial loading was less than 0.5 mm and the shear displacement in rotational loading was less than 1.0 mm (Fig. 4b).

Close

Fig. 4a Transverse displacement at the fracture site with standard deviations for movement with combined rotational and axial loading demonstrating significantly more movement with rotational shear forces with use of the intramedullary pin.

Export to PPT

Close

Fig. 4b Vertical movement at the fracture site with standard deviations for displacement with combined isolated axial loading.

Export to PPT

4 Discussion

Pilon fractures represent a complex problem both in their bony reconstruction and soft tissue management. Initially, under the aid of fluoroscopy, an indirect reduction of the fibula fracture is performed. If shortening of the fibula could not be corrected with manual manipulation, an open approach is then opted. However, particularly in the elderly population, open plating of the fibula has been shown to have complications rates as high as 18%–30 % with wound dehiscence as the most common finding.^6,7 Percutaneous fixation of associated fibular fractures offers a means to stabilize the lateral column without an added incision. Initial cohort studies demonstrate comparable functional outcomes and union rates between percutaneous and plate fixation.⁸ Biomechanical studies of nonlocked plate and intramedullary fixation have shown similar resistance to torsional loading with similar strength profiles between the two modalities.⁹ With the high rates of surgical complications from open treatment and the advance in technology of intramedullary and locked plate options for distal tibia fractures, some authors have called into question the necessity for fibular fixation as an adjunct.¹⁰

However, particularly in the simple short oblique or transverse fracture fibular fracture, anatomic reduction through closed or limited open means and percutaneous fixation allows restoration of fibular length and utilizes ligamentotaxis to facilitate articular reduction at the plafond, and continues to be pursued in the majority of cases. A myriad of techniques for percutaneous fixation have led to conflicting literature for percutaneous fibular treatment options, with complication rates ranging from 0.6 % to 8.5 %.^3,7 The wide number of treatment options for pilon and distal fibular fractures makes a heterogeneity that is difficult to evaluate with clinical studies. A majority of lab studies have evaluated comparisons between standard plate fixation with locked or unlocked fibular fixation.^4,8 However, there is a paucity of comparison of various percutaneous techniques to one another.

Our study demonstrates that both screw fixation and smooth pin fixation are effective at limiting motion at the site of a transverse fibula fracture, but with significantly less motion through intramedullary screw fixation. The larger diameter of the threaded screw and the increased endosteal purchase of the threads allow for less motion at the fracture site, providing a more stable construct for a simple fracture pattern. Although not tested in the current study, in the literature, the larger diameter of the screw allows for more stability that occurs with external rotation forces as it mechanically takes away most of the degrees of freedom at the fracture site as long as the bone quality is adequate enough to provide some fixation and purchase. More than 50 % cortical apposition has been proposed as a minimum requirement for use of an intramedullary device.¹¹ Comminution, segmental bone loss and syndesmotic damage would also weigh against the use of either a screw or nail. The bone fragments from comminution would need separate attention and bone loss could require a staged repair with grafting and plating. The use of either the screw or nail furthermore assumes that the syndesmosis continues to provide distal fibular stability. The screw or nail have nothing to offer to buttress the fibula and tibia distally against torsional loading and widening of the tibiotalar mortise.

Given similar purchases, cortical screws were chosen over cancellous screws in the current study due to providing a better diaphyseal fit similar to intramedullary nail fixation in long bone fractures. From our experience, screw purchase at the tip of the fracture should not alter healing potential compared to the overall diameter fit of the intramedullary fixation. The 4.5 mm cortical screw offered more stable fixation in torsional loading typical of daily activity; however, the amount of movement in torsion using the nail indicated that it would also suffice to produce healing clinically. In cases where the lateral skin condition or an inadequate skin bridge weigh against a separate incision for plating, either modality could offer sufficient stability. The preoperative imaging would indicate whether the larger 4.5 screw could be applied or whether the less stable smooth rod would be necessary.

Appropriate implant sizing minimizes technical difficulties associated with the screw fixation, as some fibula may not be amenable to 4.5 mm screw fixation. One specimen in our study demonstrated catastrophic failure from attempted implantation of the threaded screw with severe comminution about the fracture site occurring during screw introduction. CT studies may be used for preoperative planning and have shown a wide variety of native fibular size.¹² Adjunct fixation with use of a smooth/locked intramedullary Enders nail for the fibula has been shown to be an effective adjunct in these cases.¹³ Use of intraoperative fluoroscopy to assess the fracture site during passage of the intramedullary device is recommended to help decrease this complication.

We chose to use two loading types: (1) 2000 cycles of 5N-m external rotation torque while applying a constant 700N axial compressive force and (2) 500 cycles of 100N–700N axial compressive force. The first loading type was chosen following the approach of Switaj et al.⁴ Kim et al. also used 2000 cycles of external rotation torque, but with a magnitude of ± 1N-m.¹⁴ The axial loading was added because rotational testing under 700N axial loads is common in the literature and we believed it would be valuable to see what occurs with the unthreaded rush rod as the leg moves between a nearly unloaded state and bodyweight loading.^4,14–17

Limitations of the current study include: 1) The rod was not always fit well to the inner diameter of the canal, so in some cases it didn't do much until a certain amount of shear displacement occurred, 2) external rotation torque to failure is commonly cited in the literature and should have been part of our examination and 3) difference in implant dimensions as the screw which was 50 % larger in diameter compared to the rod, allowing for differences in shear interference and canal filling.

The current study provides a cadaveric model that closely approximates the loading represented during walking from body weight to determine which percutaneous fixation method provides more stability in treating transverse fibular fractures. We loaded and oriented in a manner that the distribution between the fibula and tibia would represent as close to physiologic as possible. Creating the boundary conditions for a specimen with a foot and tibia included under bodyweight loading provided a more accurate representation compared to creating boundary conditions to load an isolated fibula. We recommend further biomechanical studies involving larger amounts of cadaveric specimens loaded under different magnitudes and until failure.

5 Conclusion

Soft tissue management in pilon fracture treatments is essential. Both smooth pin and threaded screw fixation allowed small amounts of displacement with both shear and vertical loading in our suprasyndesmotic fibular model. Further clinical correlation is recommended, but increased stability is offered from intramedullary cortical screw fixation compared to pin fixation.

Consent

Due to the nature that this study was a retrospective chart review and no tissues or cultures were obtained to complete this study, individual informed consent was not obtained by the patients or their parents. No identifiable data is included in the manuscript or figures.

Ethics

This study was approved by the Penn State College of Medicine Institutional Review Board. This work was performed at Penn State College of Medicine and Penn State Hershey Medical Center. All laws and regulations were followed for this project.

Funding

No funding was involved. There were no proprietary interests in the materials described in the article.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Michael J. Florack: gathered and analyzed data as well as authored the manuscript. Tommy Pan: gathered and analyzed data as well as authored the manuscript, analyzed and edited the manuscript. Kevin T. Monahan: gathered and analyzed data as well as authored the manuscript, analyzed and edited the manuscript. Gregory Angelides: gathered and analyzed data as well as authored the manuscript. Patrick J. Schimoler: gathered and analyzed data as well as authored the manuscript. Mark C. Miller: gathered and analyzed data as well as authored the manuscript. Edward R. Westrick: performed the procedures, oversaw the project, edited the project, and led the research team, All authors read and approved the final manuscript. Jon E. Hammarstedt: gathered and analyzed data as well as authored the manuscript, analyzed and edited the manuscript.