See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/272843911

Anatomy and classification of the posterior tibial fragment in ankle fractures

Article  in  Archives of Orthopaedic and Trauma Surgery · February 2015

DOI: 10.1007/s00402-015-2171-4 · Source: PubMed

CITATIONS READS

63 4,901

6 authors, including:

Jan Bartonícek Stefan Rammelt

Military University Hospital Prague Universitätsklinikum Carl Gustav Carus Dresden
176 PUBLICATIONS   1,283 CITATIONS    453 PUBLICATIONS   7,143 CITATIONS   

SEE PROFILE SEE PROFILE

Karel Kostlivý Vaclav Vanecek

Thomayerova nemocnice military hospial prague
11 PUBLICATIONS   77 CITATIONS    7 PUBLICATIONS   72 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Axial and shear pullout forces of composite, porcine and human metatarsal and cuboid bones View project

Posterior malleolus fractures, Scapular fractures View project

All content following this page was uploaded by Jan Bartonícek on 07 July 2016.

The user has requested enhancement of the downloaded file.

Arch Orthop Trauma Surg (2015) 135:505–516
DOI 10.1007/s00402-015-2171-4

TRAUMA SURGERY

Anatomy and classification of the posterior tibial fragment

in ankle fractures
Jan Bartonı́ček • Stefan Rammelt • Karel Kostlivý •

Václav Vaněček • Daniel Klika • Ivo Trešl

Received: 25 June 2014 / Published online: 24 February 2015

Ó Springer-Verlag Berlin Heidelberg 2015

Abstract study. The mean patient age was 49 years (range 19–83 years).
Introduction The aim of this study was to analyze the The exclusion criteria were patients below 18 years of age,
pathoanatomy of the posterior fragment on the basis of a inability to provide written consent, fractures of the tibial pilon,
comprehensive CT examination, including 3D reconstruc- posttraumatic arthritis and pre-existing deformities. In all pa-
tions, in a large patient cohort. tients, post-injury radiographs were obtained in anteroposteri-
Materials and methods One hundred and forty one con- or, mortise and lateral views. All patients underwent CT
secutive individuals with an ankle fracture or fracture-dislo- scanning in transverse, sagittal and frontal planes. 3D CT re-
cation of types Weber B or Weber C and evidence of a posterior construction was performed in 91 patients.
tibial fragment in standard radiographs were included in the Results We were able to classify 137 cases into one of the
following four types with constant pathoanatomic features:
type 1: extraincisural fragment with an intact fibular notch,
J. Bartonı́ček (&)
V. Vaněček type 2: posterolateral fragment extending into the fibular
Department of Orthopaedic Trauma, First Faculty of Medicine, notch, type 3: posteromedial two-part fragment involving
Central Military Hospital, Charles University, U Vojenské the medial malleolus, type 4: large posterolateral triangular
nemocnice 1200, 169 02 Prague 6, Czech Republic
fragment. In the 4 cases it was not possible to classify the
e-mail: bartonicek.jan@seznam.cz
type of the posterior tibial fragment. These were collectively
V. Vaněček
termed type 5 (irregular, osteoporotic fragments).
e-mail: vaclav.vanecek@uvn.cz
Conclusion It is impossible to assess the shape and size of
S. Rammelt the posterior malleolar fragment, involvement of the fibular
University Center of Orthopedics and Traumatology, University notch, or the medial malleolus, on the basis of plain radio-
Hospital Carl Gustav Carus Dresden, Fetscherstr. 74,
graphs. The system that we propose for classification of
01307 Dresden, Germany
e-mail: stefan.rammelt@uniklinikum-dresden.de fractures of the posterior malleolus is based on CT examina-
tion and takes into account the size, shape and location of the
K. Kostlivý fragment, stability of the tibio-talar joint and the integrity of
Department of Surgery, First Faculty of Medicine and Thomayer
the fibular notch. It may be a useful indication for surgery and
Hospital, Charles University, Vı́deňská 800, 140 59 Prague 4,
Czech Republic defining the most useful approach to these injuries.
e-mail: kostlivy@seznam.cz
Keywords Ankle fractures
Posterior malleolus
D. Klika
fractures
Classifications
Pilon fractures
Department of Radiology, Central Military Hospital, U Vojenské
nemocnice 1200, 169 02 Prague 6, Czech Republic
e-mail: daniel.klika@uvn.cz
Introduction
I. Trešl
Surgical Department, Masaryk Hospital, Dukelských hrdinů 200,
269 01 Rakovnı́k, Czech Republic Fractures of the posterior rim of the distal tibia in ankle
e-mail: IvoTresl@seznam.cz fractures and fracture-dislocations have been the subject of

123
506 Arch Orthop Trauma Surg (2015) 135:505–516

continuing interest to orthopaedics and trauma surgeons for subluxation or dislocation of the talus, emergent closed
more than 100 years [1–13]. In the German literature, this reduction was carried out under sufficient intravenous
fragment is commonly referred to as Volkmanńs triangle [3, 6, analgesia and repeated radiographs were taken post re-
12], while English-speaking authors prefer the terms posterior duction and splinting. In Weber type C, injuries with a high
malleolus, posterior rim, posterior edge, or posterior fragment fibular fracture (Maisonneuve type), anteroposterior and
of distal tibia [2, 4, 8, 10]. In clinical studies, the mere pres- lateral radiographs of the lower leg were obtained.
ence of a posterior tibial fragment is a negative prognostic All patients underwent CT scanning with reconstruction
factor [14–16]. Failure to reduce the posterior tibial fragment in transverse, sagittal and frontal planes. All were required
may result in symptomatic malunion requiring correction [9, to sign informed consent. CT examination was rejected by
17]. Numerous anatomical, biomechanical and clinical studies 3 patients due to claustrophobia or other concerns and has
have been performed on this subject, often with conflicting been excluded from the study. All patients have been in-
results. As a result, the indication for open reduction and in- formed that the results of CT examination may be used for
ternal fixation of the posterior tibial fragment remains con- the study of anatomy of the posterior wall of the tibia and
troversial [12, 18–20]. So far, no generally accepted, clinically they agreed. The examinations were performed on multi-
relevant classification of these injuries exists. Several his- detector CT scanners: 64-row Somatom Sensation 64
torical classifications of fractures of the posterior malleolus are (Siemens AG, Erlangen, Germany) using the following
based on plain radiographs of the ankle. This method, how- helical study protocol: 64 9 0.6 mm, 0.55 pitch, 0° gantry
ever, has proved inadequate for a proper understanding of the tilt acquisition with 120 kV, 90 mAs, and a 512 9 512
anatomy of posterior tibial rim (Fig. 1) and the resulting matrix or 320-row Toshiba Aquilion One 320 (Toshiba
fragments [21–24]. Therefore, preference is given to com- Medical Systems Europe, Prague, Czech Republic) scanner
puted tomography (CT) examination. Friedburg et al. [25], in performing volume study in 0.5 mm section thickness in
1983, were the first to conduct a CT study of posterior mal- 160 mm range, 0° gantry tilt acquisition with 120 kV,
leolus fractures. However, only few CT-based studies have 75 mAs. Conventional axial 3.0 mm thick medium smooth
systematically addressed the anatomy and possible classifi- sections were obtained and 1.0 mm (1.0 mm increment)
cation of the posterior fragment of the distal tibia [26–28]. The sharp axial, 3.0 mm (2.0 mm increment) medium smooth
classifications presented so far have certain drawbacks and coronal and 3.0 mm (2.0 mm increment) medium smooth
discrimination between posterior malleolar and pilon fractures sagittal 2D reconstructions were reformatted from the raw
remains difficult [30]. Therefore, the aim of this study was to data with the Siemens scanner. Axial 0.5 mm bone sharp
analyze the pathoanatomy of the posterior fragment on the and soft sections were obtained with the Toshiba scanner
basis of a comprehensive CT examination, including 3D re- and 2.0 mm (2.0 mm increment) axial, coronal and sagittal
constructions, in a large patient cohort. bone sharp 2D reconstructions were reformatted. Addi-
tionally, three-dimensional volume rendering (VR) and
3–40 mm thick maximum intensity projection (MIP) re-
Patients and methods constructions were generated from the source data using
Siemens syngoMMWP VE27A (Siemens AG, Berlin,
Patients Germany) or Toshiba Vital Vitrea fX v2.1 (Toshiba Med-
ical Systems Europe, Prague, Czech Republic) workstation,
One hundred and forty one consecutive individuals (63 respectively, allowing multiprojectional display of the an-
men and 78 women) with an ankle fracture or fracture- kle. 3D CT reconstruction was performed in 91 patients
dislocation of types Weber B or Weber C and evidence of a (64.5 %).
posterior tibial fragment in standard radiographs, treated at
the authors’ departments between January 2012 and De- Evaluation
cember 2013, were included in the study. The mean patient
age was 49 years (range 19–83 years). The right side was Radiographs were used to evaluate the type of fibular
affected 83 times, and the left side 58 times. Exclusion fracture, the presence of subluxation, or dislocation, of the
criteria were patients below 18 years of age, inability to talus, the type of medial lesion, i.e., fracture of the medial
provide written consent, fractures of the tibial pilon, post- malleolus, rupture of the deltoid ligament, or the presence
traumatic arthritis and pre-existing deformities. of flake-fragment sign of the medial malleolus. A medial
clear space of 4 mm and more was classified as rupture of
Methods the deltoid ligament. The suprasyndesmal (Weber type C)
fibular fracture was classified as low (distal third), mid-
In all patients, post-injury radiographs were obtained in shaft (middle third), and high (proximal third of the
anteroposterior, mortise and lateral views. In case of fibula).

123
Arch Orthop Trauma Surg (2015) 135:505–516 507

Fig. 1 Anatomic specimen of

the right distal tibia. a Posterior
aspect displays the malleolar
groove (sulcus malleoli, SM) for
the posterior tibial tendon, the
posterior rim (PR), the posterior
tibial tubercle (PTT) and the
posterior colliculus (CP). b The
posterior tibiofibular ligament
(LTFP) inserts at the latter.
c The intermalleolar ligament
stretches from the malleolar
sulcus to that of the lateral
malleolus. d Inferior view of
ankle mortise with the posterior
rim of distal tibiae (PR) and the
posterior tibial tubercle (PTT).
e Distal tibia from lateral
aspect; 1 posterior colliculus, 2
anterior colliculus, 3
intercollicular groove with
fibers of deltoid ligament, 4
posterior tibial tubercle, 5
fibular notch, 6 anterior tibial
(Chaput́) tubercle. (Specimen
obtained from a previous
anatomical study [21])

CT scans and 3D CT were used to evaluate the shape patients with a posterior fragment bearing the whole medial
and size of the posterior tibial fragment. For this purpose, malleolus were classified as having a partial fracture of the
the posterior rim of the tibia was divided into the following tibial pilon and were therefore, excluded from the study.
parts—posterior tibial tubercle (PTT), posterior rim (PR), The two senior authors reviewed all radiographs and CT
malleolar groove (sulcus malleoli, SM) and posterior col- scans independently. In case of differing classification,
liculus (CP) (as shown in Fig. 1). Another aspect of the consensus was reached by reviewing the respective scans
evaluation was extension of the fracture into the fibular together. The cross-sectional area of the posterior fragment
notch (incisura fibularis tibiae). The intercollicular groove was expressed as a percentage of the whole cross-sectional
was set as the conventional criterion to distinguish between area of the distal tibia on CT transverse scans with the
trimalleolar fracture of the ankle and partial fracture of the respective CT software.
tibial pilon. Patients with a posterior fragment carrying The medial lesion was assessed using the classification
only the posterior colliculus were included, while those by Pankovich and Shrivram [31, 32]. Fractures were

123
508 Arch Orthop Trauma Surg (2015) 135:505–516

divided on the basis of CT into fracture of the anterior talus was recorded 29 times, 4 times posteriorly and 25
colliculus, fracture of the anterior colliculus and partial times posterolaterally.
fracture of the posterior colliculus and fracture of the whole In 43 cases the fragment was formed by the posterior tu-
medial malleolus, i.e., fracture of both colliculi. bercle only (Fig. 2b). In 20 cases the fragment consisted of the
posterior tubercle and the medial posterior rim (Fig. 4), and in
one case the fracture line extended as far as the malleolar
Results groove. By contrast, in 11 cases the fragment was small and
carried only the lateral portion of the posterior tubercle. A
In all 141 cases, the posterior fragment carried part of the quarter to one-third of the fibular notch was affected in 48 cases,
articular surface of the distal tibia. We were able to divide less than a quarter in 24 cases and half of the notch in 2 cases.
137 cases into one of the following four types with constant Transverse and sagittal CT scans showed a depressed inter-
pathoanatomic features (Fig. 2): calary joint fragment in 25 cases. The posterolateral fragment
Type 1: extraincisural fragment with an intact fibular had an average height of 17.9 mm (range 6–27 mm high) and
notch. an average anteroposterior depth of 8.7 mm (range 7–10 mm).
Type 2: posterolateral fragment extending into the The maximal transverse area of the fragment comprised 14 %
fibular notch. (range 6–20 %) of the cross-sectional area of the tibial pilon.
Type 3: posteromedial two-part fragment involving the A Weber type B fibular fracture was recorded 53 times
medial malleolus. and Weber type C fracture 21 times (12 low, 3 midshaft
Type 4: large posterolateral triangular fragment (in- and 6 high fractures). The deltoid ligament was ruptured 24
volving more than one-third of the notch). times, the medial malleolus was fractured 41 times (ante-
In the 4 cases it was not possible to classify the type of rior colliculus 2 times, anterior colliculus and part of
the posterior tibial fragment. These were collectively ter- posterior colliculus 22 times, the whole medial malleolus
med type 5 (irregular, osteoporotic fragments). 17 times); and in 9 cases no medial lesion was seen.

Type 1: extraincisural fragment Type 3: posteromedial, two-part fragment

This type was recorded in 11 (8 %) patients (8 men, 3 This type was identified in 39 (28 %) cases (13 men, 26
women) with a mean age of 53 years (range 29–77); the women) with a mean age of 44 years (range 19–82). The
right ankle was affected 7 times and the left 4 times right ankle was affected 22 times, the left one 17 times.
(Fig. 3). In the 4 cases posterolateral subluxation of the Subluxation, or dislocation, of the talus occurred 23 times,
talus occurred. In only 3 cases, the posterior tibial tubercle 7 times posterolaterally and 16 times posteriorly. A flake-
was avulsed, in 5 cases, the fragment was formed by the fragment sign (double contour of the medial malleolus)
posterior tubercle and the medial posterior rim, and in 3 was seen in the anteroposterior view in 11 cases.
cases the fracture involved only the medial posterior rim. All fragments consisted of two triangular portions of
The average height of the fragment was 11.2 mm (range different size and involved the medial malleolus (Fig. 5).
6–17 mm), and the average antero-posterior depth was The posteromedial portion typically almost always ex-
8.1 mm (range 7–10 mm). The maximal transverse area of tended more proximally than the posterolateral portion.
the fragment comprised 9 % (range 6–12 %) of the cross- The fracture line extended to the posterior colliculus in 10
sectional area of the tibial pilon. A Weber type B fibular cases, to the intercollicular groove in 28 cases, and to the
fracture was observed 7 times and Weber type C fracture 4 malleolar groove in one case. One quarter to one-third of
times (3 low fractures and one high fracture). The deltoid the fibular notch was affected in 25 cases, less than one
ligament was ruptured 8 times, the medial malleolus was quarter in 9 cases and half of the notch in 5 cases (Fig. 6).
fractured twice (once the whole medial malleolus, and once The average height of the fragment was 29.1 mm (range
the anterior colliculus with part of the posterior colliculus); 24–35 mm) and the average antero-posterior depth
and in one case no medial lesion was diagnosed. 12.7 mm (range 7–16 mm). The maximal transverse area
of the fragment comprised 24 % (range 9–34 %) of the
Type 2: posterolateral fragment cross-sectional area of the tibial pilon.
A Weber type B fibular fracture was observed 31 times,
This type was the most frequent, occurring in 74 (52 %) a Weber type C fracture 8 times (6 low and 2 high frac-
patients (41 men, 33 women), with a mean age of 50 years tures). The anterior colliculus was intact in 15 cases; the
(range 21–76). The right ankle was affected 49 times and medial clear space was less than 4 mm in 13 cases, and
the left one 25 times. Subluxation, or dislocation, of the more than 4 mm in 11 cases. The anterior colliculus

123
Arch Orthop Trauma Surg (2015) 135:505–516 509

Fig. 2 Lateral radiograph, sagittal, and axial CT scans of posterior c posteromedial two-part fragment involving the medial malleolus,
tibial fragments types. a Extraincisural fragment with an intact fibular d large posterolateral triangular fragment
notch, b posterolateral fragment extending into the fibular notch,

fractured 15 times and, in one case, the fracture involved 37.4 mm (range 30–46 mm) and the average antero-pos-
the whole medial malleolus. terior depth 18.1 mm (range 17–19 mm). The maximal
transverse area of the fragment comprised 29 % (range
Type 4: large, posterolateral triangular fragment 25–36 %) of the cross-sectional area of the tibial pilon.
A Weber type B fibular fracture of was recorded 11
This type occurred in 13 (9 %) patients (one man and 12 times, a low Weber type C fracture twice. The deltoid
women) with a mean age of 59 years (range 48–83). The ligament was ruptured twice, the medial malleolus was
right ankle was affected 8 times, the left one 5 times. fractured 10 times (the anterior colliculus with part of
Subluxation, or dislocation, of the talus occurred 11 times, posterior colliculus 3 times, the whole medial malleolus 7
4 times posterolaterally and 7 times posteriorly. times); and no medial lesion was seen in one case.
The posterior tibial fragments displayed a triangular
geometry (Fig. 7). The fragment involved the posterior Type 5: irregular osteoporotic fracture
tubercle and the medial posterior rim in 9 cases; the mal-
leolar groove in 2 cases and the posterior colliculus in 2 In 4 (3 %) women with a mean age of 70 years (range
cases (Fig. 8). The average height of the fragment was 58–83), it was impossible to classify the posterior malleolar

123
510 Arch Orthop Trauma Surg (2015) 135:505–516

Fig. 3 Examples of type 1

extraincisural posterior tibial
fragments. a Posterior tubercle
avulsion, b posterior tubercle
and posterior rim avulsion,
c posterior rim avulsion,
probably by pull of the
intermalleolar ligament

Fig. 4 Ankle fracture-

dislocation with a type 2
posterolateral fragment
extending into the malleolar
sulcus. CT scans reveal an
intermediate impacted
fragment. a Lateral view, b CT
sagittal scan, c CT transversal
scan, d 3D CT reconstruction,
posterior view (b–d)

fracture using the above-mentioned criteria, despite 3D CT Overall, the male to female ratio decreased and patient
reconstruction. The reason was a considerable comminu- age increased with classification (Table 1). The transverse
tion of fragments caused, most probably, by osteoporosis. area, involvement of the fibular notch as expressed by the
All the patients were older women, with a Weber type B fragment length ratio, and the average height and depth of
and low Weber type C fibular fracture in two cases each. the fragments increased with increasing severity of the
The deltoid ligament was ruptured once; the medial mal- fracture as expressed by the proposed classification
leolus was fractured in 3 cases. (Table 2), (Fig. 9).

123
Arch Orthop Trauma Surg (2015) 135:505–516 511

Fig. 5 A fracture-dislocation
with a type 3 two-part posterior
tibial fracture involving medial
malleolus. a, b Ap and lateral
radiographs, c, d 3-D CT scans,
e–f 2-D CT scans

Fig. 6 Four examples (a–d) of type 3 two-part posterior tibial fractures

Discussion fracture-dislocations of the ankle [33]. Open reduction and

internal fixation of the displaced posterior fragment re-
Fractures of the posterior rim of the tibia occur in ap- stores the articular surface of the distal tibia, stability of the
proximately 46 % of Weber types B and C fractures and tibiotalar joint, and integrity of the fibular notch [3, 6, 11,

123
512 Arch Orthop Trauma Surg (2015) 135:505–516

Fig. 7 Ankle fracture-dislocation with a type 4 large triangular posterolateral fragment with involvement of posterior colliculus. a Lateral
radiograph, b sagittal CT scans, c transversal CT scan, d 3D CT reconstruction, lateral aspect, e 3D CT reconstruction, posterior aspect

15, 19, 34, 35]. It also facilitates reduction of the distal was recorded in 52 % of type-I, 54 % of type-II and
fibula into the fibular notch and restores stability of the 12.5 % of type-III fragments. However, only transverse CT
tibiofibular syndesmosis [11, 19, 36, 37]. Inappropriate scans were obtained. The injury to medial structures, the
reduction of the posterior tibial fragment may result in extent of involvement of the fibular notch, and the height of
symptomatic malunion requiring corrective osteotomy [9, the fragments was not assessed. In two cases of Type-II,
17]. Nevertheless, the significance of posterior malleolar the posterior fragment included the whole medial mal-
fractures remains controversial [12, 18–20]. One of the leolus, which according to our criteria would be considered
main reasons is the absence of a generally accepted clas- tibial pilon fractures.
sification of these injuries and criteria for internal fixation. Klammer et al. [28] further divided the fractures into
Our study aimed at providing a comprehensive, easy to use three types with medial extension (Haraguchi et al. [27]
classification of posterior malleolar fractures that is useful type-II) which they termed ‘‘posterior pilon fractures’’. A
in determining the need for internal fixation of the frag- single medially-based fragment was classified as type 1
ments and the choice of surgical approach. (although not observed by the authors), a split posterior
Radiological classifications [37–40] have been based on fragment as type 2 and a fracture anterior to the posterior
the size of the fragment in the lateral view or, more colliculus as type 3. Syndesmotic disruption was added as a
specifically, on the extent of involvement of the articular modifier resulting in 6 groups. The authors conceded that
surface of the distal tibia. A number of studies have shown with just 11 cases overall they observed several ‘‘groups’’
inadequacy of radiographs in evaluation of the shape and just once. The term ‘‘posterior pilon fracture’’ was criti-
size of the posterior fragments [22–24]. Exact assessment cized by others because axial forces and articular impaction
of the size and shape of the posterior fragments requires were not addressed and the distinction to high-energy pilon
both transverse and sagittal CT scans. Fragment anatomy fractures remained unclear [30].
cannot be understood fully without 3D CT reconstructions. With 137 patients, we report one of the largest patient
So far, only few studies have assessed fractures of the cohorts with fractures of the posterior tibia. Type 1 of our
posterior tibial rim with CT scanning. Haraguchi et al. [27] proposed classification is extraincisural and corresponds to
analyzed 57 posterior malleolar fractures. The authors the Haraguchi et al. [27] type-III (‘‘small-shell’’ frag-
identified three types of fractures: a posterolateral oblique ments). These rare fractures result from avulsion of either
fragment (type-I, 67 %). a similar fragment with medial the attachment of the posterior tibiofibular ligament, or of
extension (type-II, 19 %). and a small, shell-like fragment the intermalleolar ligament [21].
(type-III, 14 %). The posterior malleolar fragment com- Type 2 of our proposed classification is a variable pos-
prised 12 % of the cross-sectional area of the tibial plafond terolateral fragment carrying 14 % of the maximal cross-
for type-I and 30 % for type-II. It was not measured for sectional area of the distal tibia usually involving one
type-III. Posterior subluxation, or dislocation, of the talus quarter to one-third of the fibular notch. This type had the

123
Arch Orthop Trauma Surg (2015) 135:505–516 513

Fig. 8 Four examples of large posterior tibial fragments with involvement of posterior colliculus, d partial pilon fracture with the
variable size. a Type 4 with involvement of posterior tubercle and posterior fragment bearing the whole medial malleolus. The latter was
rim, b type 4 with involvement of malleolar sulcus, c type 4 with therefore excluded from this study

Table 1 Overall injury pattern Type Common feature Frequency Patient M:F W-B W-C MM DL rupture
N (%) Age (years) Fx. Fx. Fx.

1 Extraincisural 11 (8 %) 53 8:3 7 4 2 8
2 Posterolateral 74 (52 %) 50 41:33 53 21 41 24
M:F male:female, W Weber, 3 Two-part, medial mall# 39 (28 %) 44 13:26 31 8 16 13
MM medial malleolus, DL 4 Large triangular 13 (9 %) 59 1:12 12 1 10 2
deltoid ligament, mall# 5 Irregular, osteoporotic 4 (3 %) 70 0:4 2 2 3 1
malleolus

highest proportion of Weber type C fibular fractures in- size of the fragments in their study is comparable to ours,
cluding the only cases of high fibular (Maisonneuve) carrying 30 and 24 % of the cross-sectional area of the
fractures observed in our study. tibial pilon, respectively. A large posterior fragment cor-
Type 3 of our proposed classification consists of a two- responding to this type was described by Felsenreich [3] as
part fragment, always involving the medial malleolus. early as in 1936 and Friedburg et al. [25] published an axial
Most probably it results from a combination of compres- CT scan of a two-part posterior fragment as early as in
sive and avulsion forces. It corresponds to type-II of the 1983. Weber [34] observed 10 of these types. As in our
Haraguchi et al. classification (‘‘medial extension’’). The study, the posteromedial part was always larger than the

123
514 Arch Orthop Trauma Surg (2015) 135:505–516

Table 2 Anatomical features

Type N Sublux Transv area Notch Notch involvement Notch involvement Notch involvement Height Depth
(%) (%) intact \1/4 1/4–1/3 1/3–1/2 (mm) (mm)

1 11 36 9 11 – – – 11.2 8.1
2 74 39 14 – 24 48 2 17.9 8.7
3 39 59 24 – 9 25 5 29.1 12.7
4 13 85 29 – – – 13 37.4 18.1
Sublux subluxation, Transv Transverse

90
Analysis of sagittal CT scans revealed that all fragments,
80 including the extraincisural ones, carried part of the ar-
70 ticular surface of the distal tibia and frequently additional,
60 Subluxation (%)
depressed intercalary joint fragments could be seen within
50 Transverse Area (%)
40 Heigth (mm)
the main fracture line (see Fig. 4).
30 Depth (mm) The number of cases with subluxation, or dislocation, of
20 the talus, the cross-sectional area of the fragment, the
10
height of the fragment, and the extent of involvement of the
0
1 2 3 4 fibular notch increased throughout the classification
Type of posterior malleolus fracture groups. This indicates that the proposed types indeed rep-
resent a scale of increasing injury severity. Like in the CT-
Fig. 9 Anatomical features. See Table 2
based study by Yao et al. [29], we found a correlation
between the length and area of the fragments.
posterolateral part and almost always included the posterior There is no clear dividing line between ankle fractures
colliculus of the medial malleolus. Anteroposterior radio- involving the tibial plafond and tibial pilon fractures be-
graphs showed a double contour (flake-fragment sign) of cause the former may result from a combination of axial
the medial malleolus in 9 out of 10 cases. We have noted and rotational forces and the definition is a question of
this sign in only 6 out of 30 cases. Klammer et al. [28] convention. Haraguchi et al. [27] used the transmalleolar
recorded 11 cases of two-part fractures and termed them as line, although they included two cases where the posterior
posterior pilon fractures. fragment carried the whole medial malleolus. In our study,
Type 4 of our proposed classification is characterized by the dividing line was the intercollicular groove. If the an-
a large, triangular posterolateral fragment. It represents a terior colliculus is part of the posterior fragment, the
transition to partial fractures of the tibial pilon and is most fracture is classified as a partial pilon fracture. When dis-
likely caused primarily by compressive forces. Type-I of tinguishing between an ankle and pilon fractures, the
the classification by Haraguchi et al. [27] combines the course of the medial malleolar fracture should also be
patterns of type 2 and type 4 of our classification. However, considered. In Weber type B and C ankle fractures, the
we believe that these are different entities. Type 4 sig- medial malleolus is fractured horizontally, or slightly
nificantly differs from type 2 with respect to fragment size obliquely. If it runs vertically to the tibial shaft, it is rather
as measured by height, depth, involvement of the fibular a partial fracture of the pilon which should also be con-
notch and transverse area as well as patient demographics. sidered in stage 2 supination adduction (Weber type A)
All but one patient were women with a mean age of fractures that display an impaction of the medial tibial
61 years, while in types 1–3 the patients were in the age plafond [19].
range of 46–49 years. Furthermore, type 4 fragments were The indications for open reduction and internal fixation
almost exclusively accompanied by Weber type B fibular of posterior malleolar fractures are also subject to discus-
fractures while type 2 fragments had the highest proportion sion [10, 12, 13, 15, 16, 18–20]. While there is no higher
of Weber type C fractures. In contrast to Haraguchi et al. level of evidence, we believe that it is essential to consider
[27], we found a correlation between the classification not only the size and displacement of the posterior frag-
types, percentage of the maximal cross-sectional area of ment, but also integrity of the fibular notch and involve-
the fragments. ment of the medial malleolus. In the authors’ practice,
In contrast to the Müller/AO and Heim classifications, fixation of the posterior fragment is not indicated in type 1
we did not observe any extraarticular fragment [37–40]. fractures but indicated in all cases of type 4 fractures. In

123
Arch Orthop Trauma Surg (2015) 135:505–516 515

fractures of types 2 and 3 the decision to operate is based 5. Nelson MC, Jensen NK (1940) The treatment of trimalleolar
individually according to the above criteria. fractures of the ankle. Surg Gynecol Obstet 71:509–514
6. Weber BG (1966) Die Verletzungen des oberen Sprunggelenkes.
The proposed classification may also be useful in de- Huber, Bern, p 102
termining the best surgical approach. In types 2 and 4, a 7. Harper MC (1990) Talar shift. The stabilizing role of the medial,
modified posterolateral approach with mobilization of lateral and posterior ankle structures. Clin Orthop Rel Res
peroneal tendons was used almost exclusively. Open re- 250:177–183
8. Karachalios T, Roidis N, Karoutis D, Bargiotas K, Karachalios
duction and fixation of the posterior malleolus is performed GG (2001) Trimalleolar fracture with a double fragment of the
first, followed by internal fixation of the fibula, either with posterior malleolus: a case report and modified operative ap-
a posterior anti-glide plate or a lateral neutralization plate. proach to internal fixation. Foot Ankle Int 22:144–149
For fixation of type 3 fragments a posteromedial ap- 9. Weber M, Ganz R (2003) Malunion following trimalleolar frac-
ture with posterolateral subluxation of the talus—reconstruction
proach is preferred. It allows reduction and fixation of the including the posterior malleolus. Foot Ankle Int 24:338–344
posterior fragment and the medial malleolus. For this type, 10. Fitzpatrick DC, Otto JK, McKinley TO, Marsh JL, Brown TD
Weber used a double posterior approach [34]. Klammer (2004) Kinematic and contact stress analysis of posterior mal-
et al. [28] used a posterolateral approach, but due to dif- leolus fractures of the ankle. J Orthop Trauma 18:271–278
11. Miller AN, Carroll EA, Parker RJ, Helfet DL, Lorich DG (2010)
ficulties in reduction of the posteromedial part added a Posterior malleolar stabilization of syndesmotic injuries is
posteromedial approach in 2 of 11 cases. Based on our equivalent to screw fixation. Clin Orthop Relat Res
experience, a single posteromedial approach is sufficient to 468:1129–1135
visualize and reduce both posterior fragments and the 12. Heim D, Niederhauser K, Simbray N (2010) The Volkmann
dogma: a retrospective, long-term, single-center study. Eur J
medial malleolus. Trauma Emerg Surg 36:515–519
In the present study, none of the type 1 fragments was 13. Gardner MJ, Streubel PN, McCormick JJ, Klein SE, Johnson JE
fixed surgically. Fixation was performed in 26 patients (2011) Surgeon practices regarding operative treatment of pos-
(36 %) with type 2 fragments, always from a posterolateral terior malleolus fractures. Foot Ankle Int 32:385–393
14. Zenker H, Nerlich M (1982) Prognostic aspects in operated ankle
approach. Surgery was indicated if a larger fragment was fractures. Arch Orthop Trauma Surg 100:237–241
displaced by more than 3 mm. Fixation was performed in 15. Jaskulka RA, Ittner G, Schedl R (1989) Fractures of the posterior
25 patients (64 %). A posteromedial approach was used in tibial margin: their role in the prognosis of malleolar fractures.
24 cases and a posterolateral approach in one case. All type J Trauma 29:1565–1570
16. Tejwani NC, Pahk B, Kenneth AE (2010) Effect of posterior
4 posterior fragments were fixed, 12 via a posterolateral malleolus fracture on outcome after unstable ankle fracture.
approach and one via a posteromedial approach. J Trauma 69:666–669
In conclusion, it is impossible to assess the shape and 17. Rammelt S, Marti RK, Zwipp H (2013) Joint-preserving os-
size of the posterior malleolar fragment, involvement of the teotomy of malunited ankle and pilon fractures. Unfallchirurg
116:789–796
fibular notch, or the medial malleolus, on the basis of plain 18. van den Bekerom MPJ, Haverkamp D, Kloen P (2009) Biome-
radiographs. The system that we propose for classification chanical and clinical evaluation of posterior malleolar fractures.
of fractures of the posterior malleolus is based on CT ex- A systematic review of the literature. J Trauma 66:279–284
amination and takes into account the size, shape and lo- 19. Rammelt S, Heim D, Hofbauer LC, Grass R, Zwipp H (2011)
Problems and controversies in the treatment of ankle fractures.
cation of the fragment, stability of the tibio-talar joint and Unfallchirurg 114:847–860
the integrity of the fibular notch. It may be a useful indi- 20. Streubel PN, McCormick JJ, Gardner MJ (2011) The posterior
cation for surgery and defining the most useful approach to malleolus: should it be fixed and why? Curr Orthop Prax
these injuries. 22:17–24
21. Bartonı́ček J (2003) Anatomy of the tibiofibular syndesmosis and
its clinical relevance. Surg Radiol Anat 25:379–386
Conflict of interest None. 22. Ferries JS, DeCoster TA, Firoozbakhsh KK, Garcia JF, Miller RA
(1998) Plain radiographic interpretation in trimalleolar ankle
fractures poorly assesses posterior fragment size. J Orthop
Trauma 12:328–331
References 23. Ebraheim NA, Mekhail AO, Haman SP (1999) External rotation-
lateral view of the ankle in the assessment of the posterior mal-
1. Chaput V (1907) Les fractures malléolaires du cou-de-pied et les leolus. Foot Ankle Int 20:379–383
accidents du travail. Masson, Paris 24. Büchler L, Tannast M, Bonel HM, Weber M (2009) Reliability of
2. Bartonı́ček J (2004) Avulsed posterior edge of tibia—Earle’s or radiologic assessment of the fracture anatomy at posterior tibial
Volkmann’s triangle? J. Bone Joint Surg 86-B:746–750 plafond in malleolar fractures. J Orthop Trauma 23:208–212
3. Felsenreich F (1931) Untersuchung über die Pathologie des so- 25. Friedburg H, Hendrich V, Wimmer B, Riede UN (1983) Com-
genannten Volkmannschen Dreiecks neben Richtlinien moderner putertomographie bei komplexen Sprunggelenksfrakturen. Radi-
Behandlung schwerer Luxationsfrakturen des oberen Sprungge- ologie 23:421–425
lenkes. Arch Orthop Unfall-Chir 29:491–529 26. Magid D, Michelson JD, Ney DR, Fishman EK (1990) Adult
4. Lauge-Hansen N (1948) Fractures of the ankle. Arch Surg ankle fractures: comparison of plain films and interactive two-
56:259–317 and three dimensional CT scans. Am J Radiol 154:1017–1023

123
516 Arch Orthop Trauma Surg (2015) 135:505–516

27. Haraguchi N, Haruyama H, Toga H, Kato F (2006) Pathoanatomy evaluation of one-year group of patients. Acta Chir Orthop Tra-
of posterior malleolar fractures of the ankle. J Bone Joint Surg matol Čech 69:243–247
Am 88-A:1085–1092 34. Weber M (2004) Trimalleolar fractures with impaction of the
28. Klammer G, Kadakia AR, Joos DA, Seybld JD, Espinosa N posteromedial tibial plafond: implications for talar stability. Foot
(2013) Posterior pilon fractures: a retrospective case series and Ankle Int 25:716–727
proposed classification system. Foot Ankle Int 34:189–199 35. Xu H, Zhang D, Fu Z, Wang T, Zhang P, Jiang B, Shen H, Wang
29. Yao L, Zhang W, Yang G, Zhu Y, Zhai Q, Luo C (2014) Mor- G, Wang G, Wu X (2012) A retrospective study of posterior
phologic characteristics of the posterior malleolar fragment: a malleolus fractures. Int Orthop 36:1929–1936
3-D computer tomography based study. Arch Orthop Trauma 36. Gardner MJ, Brodsky A, Briggs SM, Nielson JH, Lorich DG
Surg 134:389–394 (2006) Fixation of posterior malleolar fractures provides greater
30. Chen DW, Bing L, Yang YF, Yu GR (2013) Posterior pilon syndesmotic stability. Clin Orthop Rel Res 447:165–171
fractures. Foot Ankle Int 34:766–767 37. Müller ME, Nazarian S, Koch P, Schatzker J (1987) The compre-
31. Pankovich AM, Shivaram MS (1979) Anatomical basis of hensive classification of fractures of long bones. Springer, Berlin,
variability in injuries of the medial malleolus and the deltoid pp 180–191
ligament. I. Anatomical studies. Acta Orthop Scand 38. Müller ME, Allgöwer M, Schneider R, Willeneger H (eds) (1991)
50:217–223 Manual der Osteosynthese, 3rd edn. Berlin, Springer, pp 595–612
32. Pankovich AM, Shivaram MS (1979) Anatomical basis of vari- 39. Heim U (1982) Indikation und Technik der Stabilisierung des
ability in injuries of the medial malleolus and the deltoid liga- hinteren Kantendreiecks nach Volkmann ber Malleolarfrakturen.
ment. II. Clinical studies. Acta Orthop Scand 50:223–236 Unfallheilkunde 85:388–394
33. Jehlička D, Bartonı́ček J, Svatoš F, Dobiáš J (2002) Fracture- 40. Heim U (1989) Trimalleoar fractures: late results after fixation of
dislocations of the ankle in adults. Part I: epidemiological the posterior fragment. Orthopedics 12:1053–1059

123

View publication stats