Foot & Ankle International - September 2020 - 1160

1160

Figure 2. Medial view of lower leg mounted in testing
apparatus.

foot was fixed was fastened to an X-Y translation platform
incorporating low-friction linear ball bushings that allowed
transverse forces to be applied accurately to the foot via
cables, pulleys, and weights. The resulting ankle joint translations were measured directly from the platform's motion
using 2 digital displacement indicators (Model 13-002-1;
SPI) with a resolution of 0.01 mm, one oriented parallel to
the limb's sagittal plane to document anterior and posterior
drawer and one oriented in the coronal plane to measure
medial and lateral drawer (Figure 2).
To prevent subtalar joint motion from influencing the
ankle joint motion measurements, the subtalar joint was
fixed rigidly with a 3-mm screw before testing began.
During testing, the axial load applied across the ankle joint
was the sum of the mass of the specimen above the ankle
joint and the mass of the previously described linear/rotary
bearing carriage. This load averaged 50 N and was constant
during the ankle motion quantification.
A series of motion-provoking forces were applied to the
ankle joint under each operative condition, as follows:
Relative displacement (drawer) of the talus with respect to
the tibial plafond in the transverse plane was obtained by
sequentially applying a force of 67 N to the hindfoot translation platform in the anterior, posterior, lateral, and medial
directions. During these tests, tibial rotation was locked in
neutral position, and rods attached to the stationary testing
frame were positioned to bear against the lower tibia shaft
in order to prevent its deflection. Tibia shaft rotation was

Foot & Ankle International 41(9)
then re-enabled, and torsional moments of 4 Nm were
applied sequentially to the tibia shaft to provoke internal
and external rotation of the stationary talus relative to the
tibia. Linear motion of the hindfoot X-Y translation platform was unconstrained during the rotation testing. The
forces applied to the limb to provoke tibiotalar joint translation and rotation were determined in preliminary testing to
be sufficient to reproducibly translate and rotate the joint to
firm, discrete motion end points imposed by the soft tissue
and bony constraints. The initial joint position from which
each measurement was obtained was that naturally assumed
due to the applied axial load and the congruity of the ankle
joint's articular surfaces.
Each specimen was tested under 5 different operative
scenarios in the following order: (1) intact specimen with
all ligaments retained; (2) anterior and posterior syndesmosis sectioned and interosseous membrane sectioned to a
point 10 cm above the ankle joint and both the deep and
superficial components of the deep deltoid were sectioned;
(3) isolated syndesmosis fixation; (4) isolated deltoid ligament repair; and (5) both the deltoid ligament and syndesmosis repaired. The deltoid was sectioned with a scalpel
close to the insertion site of the deltoid ligament on the
medial malleolus. The syndesmotic ligaments were sectioned with a scalpel as well. Although bimalleolar equivalent ankle fractures include a lateral malleolus fracture by
definition, for this study a fibula fracture was not created.
This eliminated the variables of fragment alignment and
fixation stability, producing a "best case" scenario in which
to precisely quantify tibiotalar joint motion. Before creating
the syndesmotic injury, a foot and ankle fellowship-trained
surgeon prepared a screw hole for syndesmotic reduction.
Syndesmotic fixation was achieved with a quadricortical
3.5-mm screw inserted through the previously drilled hole.
A clamp was not used when drilling for the syndesmosis or
placing the syndesmotic screw. The ankle was held in neutral when drilling for and placing the syndesmotic screw.
Care was taken to ensure the appropriate screw trajectory to
ensure no displacement when inserting the screw. Deltoid
repair was performed using 2 double-loaded 3.5-mm titanium Corkscrew suture anchors (Arthrex, Inc, Naples, FL),
with one anchor in each colliculus. The repair was then performed in a pants over vest fashion capturing both the deep
and superficial layers of the deltoid ligament and suturing
the repair as tight as possible.

Statistical Analysis
Displacement and rotation measurements from the 9 specimens were compared among the 5 operative conditions.
Shapiro-Wilk tests were performed to ensure normal distributions of the collected data. Analysis of variance was performed for each test (anterior, posterior, lateral, and medial
drawers, internal and external rotation) separately, and

Foot & Ankle International - September 2020

Table of Contents for the Digital Edition of Foot & Ankle International - September 2020