Foot & Ankle International - September 2020 - 1070

1070

Foot & Ankle International 41(9)

Table 2. Follow-up Comparisons Stratified by Patients Who Experienced Failed Syndesmotic Screws vs Those Who Did Not.
Screw failure
(n = 56)
Follow-up, wk, mean ± SD

Control
(n = 113)

P value

53 ± 38

43 ± 32

.06

11 (20%)

14 (12%)

.25

Final dorsiflexion ROM, degrees, mean ± SD

17 ± 8

14 ± 7

.070

Final plantarflexion ROM, degrees, mean ± SD

34 ± 9

31 ± 8

.12

Maryland foot score, mean ± SD

89 ± 13

85 ± 17

.27

Removal of screws, n (%)

Abbreviation: ROM, range of motion.

Discussion
In this study, we aimed to describe the patterns of SS failure
compared with intact SS, and discern any potential implications toward reoperations or clinical function. A few characteristics and patterns of SS failure were identified in this
study sample; however, there did not appear to be a strong
association between SS failure and need for removal of
screws or postoperative ankle ROM and clinical outcome
scores.
There were a few patient and injury characteristics that
were associated with SS failure. In our sample, SS failure
was more likely to be observed in younger, male patients.
There was no significant association with ankle fracture
classification (Lauge-Hansen or OTA) or higher-energy
injuries. To our knowledge, there is no available literature
regarding the rate of syndesmotic screw breakage and backout stratified by ankle fracture classification. It is possible
that younger patients, who may have more flexible ankle
ROM and are more active compared with older patients,
have increased syndesmotic motion with ambulation postoperatively, which may lead to screw breakage.
The anatomic location of syndesmotic screw breakage
has not been studied in the past. Our data demonstrate that
syndesmotic screws most likely break in the tibial metaphysis and least likely to break in the incisura. This may possibly be related to the distance from the center of rotation for
syndesmotic motion in the incisura, which represents the
rotational lever arm; however, biomechanical evidence is
needed to support this finding. There was also a statistically
higher rate of proximal screw breakage compared with distal screw breakage in the setting of multiple-screw fixation;
however, the underlying mechanism is unclear. Nevertheless,
a prior clinical study examining the patterns of syndesmotic
screw failure did not find a difference between the rates of
screw failure if the screws were placed <2 cm from the
tibial plafond (trans-syndesmotic) compared with 2 to 5 cm
from the plafond (suprasyndesmotic).21
Other studies have investigated the association between
screw diameter and the number of syndesmotic screws and
the incidence of screw breakage.15 Stuart et al21 reported
that 3.5-mm-diameter syndesmotic screws were more likely

to break than 4- or 4.5-mm screws. In our study, all screws
used were 3.5 mm in diameter and comparisons could not
be made on screw diameter. In that study, there was no association between the number of screws placed and the rate of
screw breakage. However, a prior cadaveric study demonstrated that a single screw was at a greater risk for breaking
than 2 screws.26 Despite this evidence, our study demonstrated a higher proportion of SS failure when 2 screws
were placed compared with a single screw. While this may
not be biomechanically intuitive, it is possible that this finding may be related to uncontrolled confounding variables
such as patient activity level, degree of injury, and the presence of comorbidities such as DM. If the study sample was
much larger, a multivariate regression analysis would have
been feasible and helpful to examine the effect of these
potential confounders of the relationship between screw
number and odds of screw failure.
Debate persists regarding the clinical significance of SS
failure following trans-syndesmotic screw fixation. Our
study demonstrates that patients with screw breakage had
similar ankle plantarflexion/dorsiflexion ROM and MFS as
compared with patients with intact screws although the
sample size is likely underpowered to detect a statistical difference. Existing evidence supports better clinical outcomes
in patients with broken screws compared with patients with
intact screws.9,22 This may be predicated on multiple factors. One hypothesis is that rigid trans-syndesmotic screw
fixation creates an environment of diminished physiologic
syndesmotic motion at the ankle, which can manifest as
reduced ankle ROM and pain. Screw breakage or backout is
likely due to motion at the syndesmosis, implying more
physiologic tibiofibular kinematics.
In our study, there was a trend toward higher implant
removal rates in the failed SS group, most of which broke,
but did not back out. It is possible that this was patient perception-driven, considering that overall the clinical scores
were similar between groups. A review of the currently
available literature reveals a lack of consensus regarding
the routine removal of syndesmotic screws. Although some
studies have reported no difference in outcome between
patients with retained or removed broken screws,2,14,23,24
others found that patients with retained screws had higher

Foot & Ankle International - September 2020

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