Foot & Ankle International - July 2018 - 860

860

Foot & Ankle International 39(7)
(2) nonweightbearing active ankle dorsiflexion. Cyclic
loading from 50 N to 150 N was commenced at a rate of 1
Hz for 1000 cycles. Digital photographs were taken every
100 cycles until completion of testing. Displacement was
measured using a digital software program.18 Following
cyclic loading, a load-to-failure test was performed, at a rate
of 1000 mm per minute.

Statistical Analysis

Figure 2. A lateral view of a right foot specimen showing an
interference screw fixation of the tibialis posterior tendon into
the cuboid. Abbreviation: P. brevis, peroneus brevis.

Using load to failure as the end point, to show a mean ± SD
difference of 131 ± 81 N between groups, power analysis
(P = .05, power 0.80) suggested that the required sample
size would be 7 specimens per group.33 Since matched pairs
were not used, the sample size was increased to 12 specimens per group, a larger number than other studies.9,24,36
Two-way analysis of variance (ANOVA) was performed to
compare displacement of the tendon transfer during cyclic
loading for each technique at 10 different cycle intervals.
An independent samples t test was used to compare mean
load-to-failure forces. Data were analyzed in SPSS (SPSS
Statistics for Windows, version 24.0; SPSS, Inc, an IBM
Company, Chicago, IL), with significance set at P < .05.

Results
Cyclic Testing

Figure 3. Specimen setup.

TP tendon was identified and dissected from the muscle belly.
The proximal tendon end was held in a freeze clamp attached
to the actuator of the testing machine.30 The specimen was
positioned on the machine under the actuator so that the load
would be applied in line with the physiological line of pull of
the TP tendon.
The TP tendon was preloaded to 100 N (ramp rate 10 N/s
for 10 seconds). Baseline digital photographs were taken of
the tendon insertion point, with a ruler included to allow
subsequent scaling of the images. A pilot test using the PW
technique showed load to failure of 401 N. A previous study
had suggested that 50 N tendon tension was required to
achieve maximum ankle dorsiflexion,17 although modern
estimates suggest that the TP muscle can generate greater
force.6,7,20,37 Considering previous estimates and our pilot
results, we chose 150 N as the maximum cyclic load, which
we felt would realistically simulate 2 postoperative conditions: (1) partial weightbearing immobilized in a boot and

All 11 specimens in the PW group survived cyclical testing,
during which the mean tendon displacement was 2.9 ± 2.5
(0.02-8.9) mm. In the ISF group, 11 of 12 specimens survived cyclic testing, with a mean tendon displacement of
2.4 ± 1.1 (0.8-3.8) mm. In both groups, significant tendon
displacement occurred compared with baseline (P < .001),
but the differences between techniques were not statistically significant (P = .35). Only 1 specimen in the PW
group displaced more than 5 mm, measuring 8.9 mm at
1000 cycles. Displacement was less than 4 mm in all 11
specimens that survived cyclic testing in the ISF group.
Figure 4 shows the change in displacement, measured every
100 cycles.
In the specimen that did not survive cyclic testing
(30-year-old man), failure occurred within the first 100
cycles. The TP tendon measured 6.5 mm in diameter and a
6.5-mm pilot hole was drilled. There were no technical difficulties during screw insertion. Posttesting examination
revealed tunnel deformity, with eccentric expansion at the
site of screw insertion (Figure 5).

Load to Failure
The mean load to failure was 419.1 ± 82.6 (291.7-601.9) N
in the PW group in comparison to 499.4 ± 109.6 (333.3702.7) N in the ISF group, P = .06. The mode of failure was

Table of Contents for the Digital Edition of Foot & Ankle International - July 2018