FOOT AND ANKLE

ORTHOPEDICS April 2011;34(4):272.

Ankle Fusion Stability: A Biomechanical Comparison of External Versus Internal Fixation

by Justin R. Hoover, MD; Robert D. Santrock, MD; William C. James, MD

Abstract

This biomechanical study compares bimalleolar external fixation to conventional crossed-screw construct in terms of stability and compression for ankle arthrodesis. The goals of the study were to determine which construct is more stable with bending and torsional forces, and to determine which construct achieves more compression.

Fourth-generation bone composite tibia and talocalcaneal models were made to 50th percentile anatomic specifications. Fourteen ankle fusion constructs were created with bimalleolar external fixators and 14 with crossed-screw constructs. Ultimate bend, torque, and compression testing were completed on the external fixator and crossed-screw constructs using a multidirectional Materials Testing Machine (MTS Systems Corp, Eden Prairie, Minnesota). Ultimate bend testing revealed a statistically significant difference (P=.0022) with the mean peak load to failure for the external fixator constructs of 973.2 N compared to 612.5 N for the crossed-screw constructs. Ultimate torque testing revealed the mean peak torque to failure for the external fixator construct was 80.2 Nm and 28.1 Nm for the crossed-screw construct, also a statistically significant difference (P=.0001). The compression testing yielded no statistically significant difference (P=.9268) between the average failure force of the external fixator construct (81.6 kg) and the crossed-screw construct (81.2 kg).

With increased stiffness in both bending and torsion and comparable compressive strengths, bimalleolar external fixation is an excellent option for tibiotalar ankle arthrodesis.

Ankle arthrodesis has historically been the mainstay of surgical treatment of end-stage ankle arthrosis. Various constructs have been described to create a stable tibiotalar fusion. Open tibiotalar joint preparation with internal crossed-screw fixation is widely used. However, other options are available, and less invasive procedures are particularly useful in patients with a history of infection, poor soft tissue quality, or poor wound-healing capabilities. Open crossed screw construct, plating, intramedullary nail, and external fixation to obtain fusion are described in the literature, and many biomechanical studies have been performed.^1-13 The basic premise remains the same: no matter what type of fixation is chosen, stability and compression are needed for successful fusion. It is for this reason that the biomechanical properties of different techniques and constructs continue to be investigated with the goals of improving stability and, in turn, patient outcomes.

In this study, we compared traditional crossed-screw fixation to bimalleolar external fixation in bending, torsion, and compression. Bimalleolar external fixation was originally described by Charnley in 1951.⁵Since that time, use of these constructs has been relatively limited. Recently, external fixation has gained interest due to the perceived stability. Our hypothesis for this study was that bimalleolar external fixation is superior to crossed-screw technique in bending strength, torsional strength, and compression.

Materials and Methods

Fourth-generation bone composite replicated tibia and talocalcaneal constructs were made to 50th percentile human anatomic specifications.^14,15 These constructs have been shown to be comparable to natural bone in terms of their biomechanical properties.¹⁴ Each construct was cut precisely to replicate the average human ankle and foot size, therefore giving the appropriate lever arms during mechanical testing. The tibial construct was 40×40×180 mm. The talocalcaneal foot construct block was 40×60×173 mm with a trapezoid talar dome 10 mm in height and a tibiotalar contact surface of 40×25 mm.

A total of 28 constructs were built for mechanical testing. Fourteen external fixation constructs were created using Sidekick Stealth bimalleolar external fixators (Wright Medical Technology, Inc, Arlington, Tennessee). Fourteen conventional crossed-screw fixation constructs were created using 6.5-mm Darco partially threaded cannulated screws (Wright Medical Technology, Inc).

To create the external fixator constructs, four 4.0×300-mm transfixing external fixator pins were inserted in standard fashion in the constructs’ talar neck, calcaneus, and proximal and distal tibia. The talar neck pin was placed 25 mm plantar to the talar dome surface and 15 mm anterior. The calcaneal pin was placed 15 mm proximal to the plantar surface and 15 mm anterior to the posterior aspect of the calcaneal block. The distal tibial pin was placed 30 mm proximal to the construct plafond and 12 mm posterior to the anterior tibial crest. The proximal tibial pin was placed 90 mm proximal to the construct plafond and 20 mm posterior to the anterior tibial crest. The medial and lateral Sidekick Stealth external fixator frames were then placed 32 mm from the medial and lateral edge of the composite bone (Figure 1).

Figure 1: Bimalleolar external fixator construct. Figure 1: Bimalleolar external fixator construct.

The internal fixation crossed-screw constructs were built by predrilling the tibial and talocalcaneal constructs with a 4.4-mm drill bit. This was followed by countersinking the tibial composite bone. Two 6.5×70-mm Darco partially threaded headed screws were placed in standard crossed screw construct from the medial and lateral side of the tibia construct into the talocalcaneal construct (Figure 2).

Ultimate Bend Testing

Four external fixator and 4 crossed-screw constructs were fashioned in the manner outlined above. The tibial component was secured to the base plate box of the multidirectional Materials Testing Machine (MTS machine; MTS Systems Corp, Eden Prairie, Minnesota). The actuator was placed on the plantar surface of the talocalcaneal construct 12.7 mm from the distal end. The MTS machine actuator applied a load at 1 mm per second with data acquisition at 50 Hz until failure. Data was recorded in newtons of force. Failure was defined as a drop in the load (caused by composite fracture, screw pullout, or brace plate fracture).

Ultimate Torque Testing

Four external fixator and 4 crossed-screw constructs were fashioned in the manner outlined above. The tibial component was secured to the base plate box of the multidirectional MTS machine. The actuator was placed on the plantar surface of the talocalcaneal construct in line with the center of the talar dome. The MTS machine actuator applied a load of 1° per second with data acquisition at 25 Hz until failure. Data was recorded in newton-meters. Failure was defined as a drop in the load (caused by composite fracture, screw pullout, or brace plate fracture).

Compression Testing

Six external fixator and 6 crossed-screw constructs were fashioned in the manner outlined above with the exception of placing a 6.35-mm Load Washer Load Cell (Interface, Inc, Scottsdale, Arizona) and a 76.2×76.2×0.8-mm aluminum plate between the tibial and talocalcaneal composite bone (Figure 3). The external fixator talocalcaneal pins were tightened in a static position both medially and laterally. The tibial component was advanced proximally by turning each of the 4 nuts one-quarter turn. Measurements at each one-quarter turn were made in kilograms of force. This was completed until failure. The crossed-screw compression was measured similarly. The crossed screws were set when the underside of the screw head contacted the composite bone material. Once this contact was made, each screw was tightened one-quarter turn, and kilograms of force measurements were made after each one-quarter turn until failure. The number of turns and kilograms of force were recorded. Failure was defined as fracture of the composite bone material in the crossed-screw construct and damage to the articulating bar threads leading to inability to advance the nuts on the medial and lateral braces in the external fixator group.

Figure 3: Compression testing model with transducer.

It is to be noted that only 5 external fixator constructs were analyzed, as external fixator sample 1 was loaded without the aluminum plate and therefore was not equivalent to the other constructs and was eliminated from the data. The aluminum plate was necessary to insert because the ring transducer embedded into the composite bone when compressed.

Statistical Analysis

The mean, maximum, minimum, and standard deviations of each parameter at failure were calculated. Student t test was used to compare the external fixation group to the crossed-screw construct group. Statistical significance level was set at P<.05.

Results

The mean peak load to failure for dorsiflexion for the external fixator was 973.2±109.5 N and for the crossed-screw construct was 612.5±89.2 N, which was a statistically significant difference (P=.0022) (Table 1, Figure 4).

Figure 4: Peak load (n=4).

The mean bending stiffness of the composite tibiotalar joint for each condition is shown in Table 2. The Stealth external fixator yielded a bending stiffness of 34,373±2017 Nm, and the Darco crossed-screw construct was 26,285±1416 Nm, which is a statistically significant difference (P=.0006) (Table 2, Figure 5).

Figure 5: Bending stiffness (n=4).

Ultimate torque testing revealed the mean peak torque to failure for the external fixator construct was 80.2±11.5 Nm and the crossed-screw construct was 28.1±3.0 Nm. This was a statistically significant difference (P=.0001) (Table 3, Figure 6).

Figure 6: Mean peak torque (n=4).

The mean torsional stiffness for the external fixator construct was 3.09±0.10 Nm/degree. The crossed-screw construct mean torsional stiffness was 1.36±1.04 Nm/degree. This was a statistically significant difference using the t test (P=.0324) (Table 4, Figure 7).

Figure 7: Torsional stiffness (n=4).

Compression testing yielded no statistically significant difference (P=.9268) between the average failure force of the external fixation (81.6 kg) and the crossed screws (81.2 kg). The external fixator failure mode was the nut would not advance further due to damage of the articulating bar threads. Four turns of each nut was the average number of turns to failure. The failure mode for all Darco crossed-screw constructs was fracture of the composite bone. The average number of screw turns was 1.5 turns after the screw was seated (Table 5).

Discussion

Ankle arthrodesis is a valuable procedure for end-stage ankle arthrosis. Multiple techniques have been described, including internal and external fixation. Numerous clinical studies have shown internal crossed screw construct as advantageous due to high rates of fusion, decreased rates of infection, and improved patient comfort.^1-13 However, in many cases external fixation may be a more appropriate means of facilitating fusion. Ring external fixators have been shown to be a viable option to establish fusion and is particularly useful in cases of past infection, poor wound-healing capabilities, or decreased bone stock.¹⁰ To our knowledge, no study has compared a compression arthrodesis technique using bimalleolar external fixation to crossed screws. Our hypothesis was that bimalleolar external fixation is a more rigid construct with regard to bending strength (dorsiflexion), torsional strength, and compression than crossed screws.

Our results revealed statistically significant differences in both bending forces (P=.0022) and torsional forces (P=.0001), showing that external fixation is more rigid than crossed screws. No statistical significance was seen with regard to compression.

Our study protocol had several limitations, including the use of fourth-generation bone composite, not accounting for subtalar joint involvement, space-occupying use of the load cell ring transducer, effects of musculature and ligaments, and unidirectional force measurements performed on the MTS machine at nonphysiologic rates. Our plans include performing a cadaveric study with multidirectional testing using a bimalleolar external fixator and crossed screws. These anatomic specimens will allow us to take into account the subtalar joint. Additionally, we would like to use less invasive devices to assess compression.

Conclusion

Bimalleolar external fixator is a more rigid construct in both bending and torsion as compared to conventional crossed screws. There was no significant difference between the 2 methods with regard to compression. However, with increased stiffness in both bending and torsion and comparable compressive strengths, a bimalleolar external fixator is an excellent option for tibiotalar ankle arthrodesis in the correct clinical setting. Clinical indications would include patients with a history of infection or poor soft tissue quality.⁶ In our practice, it is also useful for arthroscopic-assisted ankle fusions due to its minimally invasive nature; it has become our primary mode of fixation for these fusions. Biomechanical testing predicts that it has a low probability of failure due to rigidity of the construct. External fixation was found to be more rigid than a traditional lag-screw technique. These findings may extrapolate into increased fusion rate and earlier or immediate weight bearing in patients undergoing ankle arthrodesis. This would potentially reduce the associated morbidity of extended nonweight bearing while increasing patient satisfaction.

For further information: http://www.orthosupersite.com/view.aspx?rid=81555