A reliable radiographic measurement for evaluation of normal distal tibiofibular syndesmosis: a multi-detector computed tomography study in adults

Syndesmotic injury may be difficult to diagnose using radiographs, because of variability
in the amount of rotation, the wide anatomic variability in the depth of the peroneal
groove, and the shape of the tibial tubercles. Quantitative parameters of the morphological
measurements should be based on a scientific and reliable method; otherwise, the measurement
results would not be credible 22]. Accuracy and precision are the most important parameters for evaluation of quantitative
measurements for an area of known variability. Traditional common radiographic views
for the tibiofibular syndesmosis include weight-bearing AP, mortise, and lateral views
of the ankle. Although a mortise view may demonstrate diastasis, it is difficult to
perform under post-traumatic conditions. CT is more sensitive than radiography for
detecting minor degrees of syndesmotic injury 12], but the selection of the observation plane for 2-D axial CT is still influenced
by the position of the ankle. SSD, also known as surface rendering, was the first
3-D rendering technique applied to medical data sets. Its early development in the
1970s was a logical extension of new computer graphic and image-processing techniques
and innovations in data segmentation and display 26], 27]. 3-D SSD rendering technology is one of the computer research priorities in the field
of visual pattern recognition. Therefore, the 3-D SSD rendering technology was used
in our study.

As the syndesmosis plays an important role in stability of the ankle, it was suggested
that magnetic resonance imaging (MRI) or CT should be used to ensure adequate diagnosis
and evaluation 21], 30]. Franke et al. 31] recommend that persistent dislocation should be verified and detected with intraoperative
3-D imaging or postoperative CT, after reduction and syndesmotic screw fixation of
unstable syndesmotic injuries. Malhotra et al. 21] suggested that a comparison of angular and area measurements would help identify
a diastasis of the syndesmosis in patients with persistent pain after a rotational
ankle fracture. They performed an angular measurement subtended by two lines drawn
tangent to the anterior and posterior surfaces of the distal tibia and lateral malleolus.
The area measured was bounded by above two lines, the lateral tibia and the medial
aspect of the lateral malleolus.

Recently, the lateral view of the ankle has been studied with regard to the relationship
of the distal tibia and fibula. Grenier et al. 32] have established a novel radiographic parameter based on the lateral view of the
ankle, i.e., the anteroposterior tibiofibular (APTF) ratio, which is normally 0.94?±?0.13
and can be used to identify and prevent a malreduction of the fibula in the incisura
fibularis. Croft et al. 33] showed that the anterior tibiofibular ratio (defined as the ratio of the tibial width
to the anterior tibiofibular interval) is reproducibly measured, and is suitable for
determining the normal relationship of the tibia and fibula. Some authors used CT
to postoperatively evaluate the reduction of the injured syndesmosis 21], 27], 34]. Parameters analogous to those traditionally measured in previous studies 18], 34], 35], including the anterior TCS, posterior TCS and others, were measured using 3-D and
2-D CT. According to these parameters, measurements of the tibiofibular syndesmosis
aided in the diagnosis of diastasis. With the increasing use of CT, the rotational
changes of the syndesmosis can be measured on axial slices. The rotation of the fibula
relative to the tibia is a common parameter; this is described by an angle between
a line tangential to the anterior and posterior tibial incisura and a line through
the anterior and posterior fibular tubercles 17], 35]. Moreover, Ebinger et al. 36] demonstrated that the rotational displacement of the fibula using a full 3-D CT model,
correlated with the average distance between anterior and posterior tibiofibular joint
space. In terms of the vertical displacement of the fibula, they indicated that the
longitudinal displacement correlated with the posterior tibiofibular joint space measurement
on axial CT imaging.

The common measuring plane of the distal tibiofibular syndesmosis on radiograph is
1 cm proximal to the tibial plafond 10], 11], 15], 16]. Elgafy et al. 18] reported that the width of the tibiofibular syndesmosis on the axial CT images was
measured in the third section, which was 3 mm thick and 9-12 mm proximal to the tibial
plafond. Due to the existence of anatomic variability, a uniform measurement plane
lacked a scientific basis. In our study, based on theoretical computational geometry,
we precisely located the tibial plafond and the lateral prominent point of the anterior
tubercle (point A, defined as the intersection of the tangential along the tibial
tubercle) on the 3-D SSD images. The measurement plane through the lateral prominent
point and parallel to the tibial plafond was then selected. The maximum TCS or TFO
and IFD were considered through the plane, on which relevant parameters were measured.
The method could help locate the measuring plane accurately, which differed by sex
and anatomic variability in 484 samples. The results showed that the sections were
12.1 mm proximal to the tibial plafond in the male group and 7.8 mm in the female
group. The parameters provided comparable measurements planes on the axial CT image
and radiographs.

A radiological study using cadavers demonstrated that normally there is a TCS of less
than 6 mm on the AP and mortise views; the TFO on the AP view is greater than 6 mm,
or 42 % of the fibular width, and greater than 1 mm on the mortise 10]. Ostrum et al. 15] reported a TCS of less than 5.16 mm in females and 6.47 mm in males on AP view radiographs;
a TFO more than 2.1 mm in females and 5.7 mm in males indicates an intact syndesmosis.
The results for the TCS described in the present study are different from those of
previously reported radiographic studies. Elgafy et al. 18] reported that the mean anterior and posterior width of the distal tibiofibular syndesmosis
in CT scans, were 2 mm and 5 mm in male, respectively, and 2 mm and 4 mm in women.
Yeung et al. 37] demonstrated that axial CT measurements of the tibiofibular distance (TFD) were useful
predictors for syndesmosis instability in fractured ankles. They measured the TFD
in both normal and fractured ankles. The mean TCS-P measured in the current study
was 3.6 mm in the male group and 2.9 mm in the female group. This was comparable to
the posterior TFD measured between the analogous reference points on axial CT imaging,
which is normally 4.1 mm 37]. In fractures involving the tibia and fibula, 3-D or 2-D measurements could be performed
as in normal ankles, when the measurement plane above the tibial plafond is previously
well defined. Meanwhile, the pre-defined osseous points are identified regardless
of the displacement. In the current study, the TCS was divided into two components
on the 3-D SSD and 2-D axial CT images, i.e., anterior and posterior. The two parameters
did not differ significantly between the 3-D and 2-D images, because the measurement
plane on the 2-D axial CT image was defined precisely by 3-D images. According to
the defined TCS, there was a significant difference between posterior TCS in the 3-D
image or 2-D axial image and the TCS on the AP view radiographs.

Although TFO was an important measurement for assessment of the syndesmosis on radiographs,
it was difficult to assess on the 2-D axial CT section or 3-D image, because the measurement
views changed with the rotation of the foot. Therefore, only the TFO on the radiograph
was measured in the present study. The accuracy and reliability of TFO would be affected,
if it was not measured on a standard AP or coronal radiographic view. The IFD of the
tibia may predispose to displacement of the fibula in association with fracture dislocation.
Ebraheim et al. 13] found that the depth varied from 1.72 to 6.78 mm with an average of 4.29 mm at the
level of 10 mm proximal to the tibial plafond. However, they pointed out that this
wide variability in the IFD of the tibia, and in the shape of the tibial tubercle
would make evaluation difficult. In our study, we measured the depth in the 3-D images
and 2-D axial images using a precisely located plane via the intersection of the tangent
along the tubercle and parallel to the tibial plafond. Therefore, we considered the
depth value to have a particular significance, which could compensate for anatomic
variability.

The limitation of our study was mainly that the software for 3-D morphological evaluation
was not commonly available. However, the 3-D SSD rendering technique based on the
data of DICOM 3.0 format is accurate; in addition, 3-D measurement technique is already
widely used in the field of engineering modeling. Second, the cases included in the
study had forefoot injuries or pathology, but not injuries of the tibiofibular syndesmotic
ligament complex, according to radiographs and MDCT scans. Third, a CT scan is necessary
for the 3-D measurement technique, which causes radiation exposure. However, technological
developments have led to a significant reduction in radiation exposure and image noise
in the latest MDCT generation 38], 39].