Advertisement

The influence of size and comminution of the posterior malleolus fragment on gait in trimalleolar ankle fractures

  • Author Footnotes
    1 Contributed equally to the study.
    S. Schoenmakers
    Footnotes
    1 Contributed equally to the study.
    Affiliations
    Department of Surgery, Division of Trauma surgery, Maastricht University Medical Center, P. Debyelaan 25, PO Box 5800, 6202 AZ Maastricht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    1 Contributed equally to the study.
    M. Houben
    Correspondence
    Corresponding author.
    Footnotes
    1 Contributed equally to the study.
    Affiliations
    Department of Surgery, Division of Trauma surgery, Maastricht University Medical Center, P. Debyelaan 25, PO Box 5800, 6202 AZ Maastricht, the Netherlands
    Search for articles by this author
  • S. van Hoeve
    Affiliations
    Department of Surgery, Division of Trauma surgery, Maastricht University Medical Center, P. Debyelaan 25, PO Box 5800, 6202 AZ Maastricht, the Netherlands
    Search for articles by this author
  • P. Willems
    Affiliations
    Department of Movement Sciences, Maastricht University Medical Center, P. Debyelaan 25, PO Box 616, 6200 MD Maastricht, the Netherlands

    NUTRIM, School for Nutrition and Translational Research in Metabolism, PO Box 616, 6200 MD Maastricht, the Netherlands
    Search for articles by this author
  • K. Meijer
    Affiliations
    Department of Movement Sciences, Maastricht University Medical Center, P. Debyelaan 25, PO Box 616, 6200 MD Maastricht, the Netherlands

    NUTRIM, School for Nutrition and Translational Research in Metabolism, PO Box 616, 6200 MD Maastricht, the Netherlands
    Search for articles by this author
  • M. Poeze
    Affiliations
    Department of Surgery, Division of Trauma surgery, Maastricht University Medical Center, P. Debyelaan 25, PO Box 5800, 6202 AZ Maastricht, the Netherlands

    NUTRIM, School for Nutrition and Translational Research in Metabolism, PO Box 616, 6200 MD Maastricht, the Netherlands
    Search for articles by this author
  • Author Footnotes
    1 Contributed equally to the study.
Open AccessPublished:December 12, 2021DOI:https://doi.org/10.1016/j.clinbiomech.2021.105550

      Highlights

      • Malleolus tertius fracture size has impact on the range of motion of the ankle.
      • Patients with trimalleolar fractures have a lower walking speed.
      • Trimalleolar ankle fractures have decreased range of motion talocrural.
      • Decreased range of motion talocrural is related to worse functional outcome.

      Abstract

      Background

      Ankle fractures involving the posterior malleolus generally lead to worse outcome. However, no studies on gait in trimalleolar ankle fractures have evaluated the influence of size and comminution of the posterior malleolar fragment.

      Methods

      We expected patients with more severely comminuted posterior malleolus, more severe fracture type and larger posterior fragment to have reduced gait kinematics and poorer patient-reported outcomes. 26 trimalleolar ankle fracture patients were compared with 14 healthy controls and kinematically analyzed using the Oxford Foot Model. Functional outcome was based on 4 patient reported outcome questionnaires. Effects of posterior fragment size, comminution and Haraguchi fracture classification were determined on conventional and 3D CT-scans.

      Findings

      Trimalleolar patients had lower walking speed and reduced range of motion between the hindfoot and tibia in both loading and push-off phases in the sagittal and transverse planes. The range between the hindfoot and tibia in the sagittal plane in the push-off phase correlated significantly with patient reported outcomes. The absolute and relative surface area of the posterior fragment on conventional CT-scans and 3D CT-scans, correlated significantly with range of motion. Patients with a posterior malleolus size >10% of the posterior malleolus had lower flexion-extension between forefoot and hindfoot during loading phase than patients with a size ≤10%.

      Interpretation

      Trimalleolar fractures reduce walking speed and range of motion in the talocrural joint. Reduced range in the talocrural joint is associated with poorer outcomes. Posterior fragment size correlated significantly with range of motion in talocrural and midfoot joints and with patient reported outcomes.
      Level of evidence: Level 3, retrospective study.

      Keywords

      1. Introduction

      Each year 1 out of 800–1000 persons suffer an ankle fracture, accounting for 9% of all fractures (
      • Broos P.L.
      • Bisschop A.P.
      Operative treatment of ankle fractures in adults: correlation between types of fracture and final results.
      ;
      • Donken C.C.
      • Al-Khateeb H.
      • Verhofstad M.H.
      • van Laarhoven C.J.
      Surgical versus conservative interventions for treating ankle fractures in adults.
      ). Seven to 44% of the ankle fractures involve a posterior malleolus fragment (PMF) (
      • De Vries J.S.
      • Wijgman A.J.
      • Sierevelt I.N.
      • Schaap G.R.
      Long-term results of ankle fractures with a posterior malleolar fragment.
      ). Numerous studies have reported worse pain and poorer functional and radiographic outcomes for ankle fractures involving a larger PMF size and step-off (
      • Broos P.L.
      • Bisschop A.P.
      Operative treatment of ankle fractures in adults: correlation between types of fracture and final results.
      ;
      • Drijfhout van Hooff C.C.
      • Verhage S.M.
      • Hoogendoorn J.M.
      Influence of fragment size and postoperative joint congruency on long-term outcome of posterior malleolar fractures.
      ;
      • Evers J.
      • Barz L.
      • Wahnert D.
      • Gruneweller N.
      • Raschke M.J.
      • Ochman S.
      Size matters: the influence of the posterior fragment on patient outcomes in trimalleolar ankle fractures.
      ;
      • Jaskulka R.A.
      • Ittner G.
      • Schedl R.
      Fractures of the posterior tibial margin: their role in the prognosis of malleolar fractures.
      ;
      • Tejwani N.C.
      • Pahk B.
      • Egol K.A.
      Effect of posterior malleolus fracture on outcome after unstable ankle fracture.
      ;
      • Xu H.L.
      • Li X.
      • Zhang D.Y.
      • Fu Z.G.
      • Wang T.B.
      • Zhang P.X.
      • et al.
      A retrospective study of posterior malleolus fractures.
      ). Currently, There is still a lot of debate about what PMFs size should be fixated at least all authors agree PMF >25–33% of the tibial plafond and/or with a step-off ≥2 mm are considered unstable. But the smaller sizes >10% are debatable. With also the shift to a more fracture morphology (
      • Blom R.P.
      • Hayat B.
      • Al-Dirini R.M.A.
      • Sierevelt I.
      • Kerkhoffs G.M.M.J.
      • Goslings J.C.
      • Jaarsma R.L.
      • Doornberg J.N.
      • EF3X-trial Study Group
      Posterior malleolar ankle fractures.
      ). approach. Hence, surgical treatment with open reduction and internal fixation of the fragment is advised, although the exact cut-off points for what therapeutic guidelines are under debate (
      • Odak S.
      • Ahluwalia R.
      • Unnikrishnan P.
      • Hennessy M.
      • Platt S.
      Management of Posterior Malleolar Fractures: a systematic review.
      ;
      • van den Bekerom M.P.
      • Haverkamp D.
      • Kloen P.
      Biomechanical and clinical evaluation of posterior malleolar fractures. A systematic review of the literature.
      ;
      • Veltman E.S.
      • Halma J.J.
      • de Gast A.
      Longterm outcome of 886 posterior malleolar fractures: a systematic review of the literature.
      ). Congruency of the tibial plafond, comminution, residual talar displacement and syndesmotic stability are also important prognostic factors, and should be considered in the decision to surgically treat a PMF (
      • Odak S.
      • Ahluwalia R.
      • Unnikrishnan P.
      • Hennessy M.
      • Platt S.
      Management of Posterior Malleolar Fractures: a systematic review.
      ;
      • Tenenbaum S.
      • Shazar N.
      • Bruck N.
      • Bariteau J.
      Posterior malleolus fractures.
      ). Furthermore, clinical decision making should be based on CT-scans, ideally 3D CT-scans, instead of X-rays, to enhance the diagnostic accuracy for the articular surface involvement of the PMF (
      • de Muinck Keizer R.O.
      • Meijer D.T.
      • van der Gronde B.A.
      • Teunis T.
      • Stufkens S.A.
      • Kerkhoffs G.M.
      • et al.
      Articular gap and step-off revisited: 3D quantification of operative reduction for posterior malleolar fragments.
      ;
      • Mangnus L.
      • Meijer D.T.
      • Stufkens S.A.
      • Mellema J.J.
      • Steller E.P.
      • Kerkhoffs G.M.
      • et al.
      Posterior malleolar fracture patterns.
      ;
      • Meijer D.T.
      • Doornberg J.N.
      • Sierevelt I.N.
      • Mallee W.H.
      • van Dijk C.N.
      • Kerkhoffs G.M.
      • et al.
      Guesstimation of posterior malleolar fractures on lateral plain radiographs.
      ;
      • Meijer D.T.
      • de Muinck Keizer R.J.
      • Doornberg J.N.
      • Sierevelt I.N.
      • Stufkens S.A.
      • Kerkhoffs G.M.
      • et al.
      Diagnostic accuracy of 2-dimensional computed tomography for articular involvement and fracture pattern of posterior malleolar fractures.
      ). Only a few studies have used gait analysis to evaluate the kinematics of the ankle joint in ankle fracture patients (
      • Becker H.P.
      • Rosenbaum D.
      • Kriese T.
      • Gerngross H.
      • Claes L.
      Gait asymmetry following successful surgical treatment of ankle fractures in young adults.
      ;
      • Belcher G.L.
      • Radomisli T.E.
      • Abate J.A.
      • Stabile L.A.
      • Trafton P.G.
      Functional outcome analysis of operatively treated malleolar fractures.
      ;
      • Losch A.
      • Meybohm P.
      • Schmalz T.
      • Fuchs M.
      • Vamvukakis F.
      • Dresing K.
      • et al.
      Functional results of dynamic gait analysis after 1 year of hobby-athletes with a surgically treated ankle fracture.
      ;
      • Segal G.
      • Elbaz A.
      • Parsi A.
      • Heller Z.
      • Palmanovich E.
      • Nyska M.
      • et al.
      Clinical outcomes following ankle fracture: a cross-sectional observational study.
      ;
      • van Hoeve S.
      • Houben M.
      • Vebruggen J.P.A.M.
      • Willems P.
      • Meijer K.
      • Poeze M.
      Gait analysis related to functional outcome in patients operated for ankle fractures.
      ;
      • Wang R.
      • Thur C.K.
      • Gutierrez-Farewik E.M.
      • Wretenberg P.
      • Brostrom E.
      One year follow-up after operative ankle fractures: a prospective gait analysis study with a multi-segment foot model.
      ). No studies have performed a biomechanical gait analysis, using a multi-segment foot model, to analyze the effects of these PMF characteristics in trimalleolar ankle fractures.
      The objective of this study was to analyze the ankle kinematics in patients with a trimalleolar ankle fracture, using the four-segment Oxford Foot Model (OFM), and relate this to the radiographic features of the fracture (including comminution, size of the PMF based on 3D CT-scans, and fracture type) (
      • van Hoeve S.
      • de Vos J.
      • Weijers P.
      • Verbruggen J.
      • Willems P.
      • Poeze M.
      • et al.
      Repeatability of the Oxford foot model for kinematic gait analysis of the foot and ankle.
      ;
      • Wright C.J.
      • Arnold B.L.
      • Coffey T.G.
      • Pidcoe P.E.
      Repeatability of the modified Oxford foot model during gait in healthy adults.
      ). Radiographic characteristics, ankle kinematics, and questionnaires were analyzed.
      Our hypothesis was that patients with more severely comminuted posterior malleolus, a more severe fracture type and larger PMFs would have reduced gait kinematics and worse patient-reported outcomes (PROMs). In addition, we hypothesized that the RoM between the hindfoot and tibia would be significantly lower during walking in patients with a PMF, compared to healthy controls.

      2. Methods

      2.1 Inclusion

      This retrospective cohort study included 26 patients with a trimalleolar ankle fracture. All patients had an ankle fracture with a PMF, had undergone a CT-scan of the fractured ankle and had had surgical treatment. Patients were included at least 6 months after surgical treatment, and were followed for one year (
      • Agostini V.
      • Ganio D.
      • Facchin K.
      • Cane L.
      • Moreira Carneiro S.
      • Knaflitz M.
      Gait parameters and muscle activation patterns at 3, 6 and 12 months after total hip arthroplasty.
      ;
      • Beckenkamp P.R.
      • Lin C.W.
      • Chagpar S.
      • Herbert R.D.
      • van der Ploeg H.P.
      • Moseley A.M.
      Prognosis of physical function following ankle fracture: a systematic review with meta-analysis.
      ;
      • Nantel J.
      • Termoz N.
      • Vendittoli P.A.
      • Lavigne M.
      • Prince F.
      Gait patterns after total hip arthroplasty and surface replacement arthroplasty.
      ). The patients with trimalleolar fractures were compared with 14 healthy controls (26 ft; 2 patients not bilateral due to data loss). In both groups, the following exclusion criteria were used: age < 18 or > 80 years, contralateral or ipsilateral fractures or neurotrauma, pathological fractures, pre-existent abnormalities and spinal or neurological injury of the lower extremities.

      2.2 Protocol

      Four questionnaires were used for PROM evaluation: Short Form-36 (SF-36), Foot & Ankle Disability Index (FADI), American Orthopaedic Foot Ankle Score (AOFAS), and Visual Analogue Scale (VAS) (
      • Carlsson A.M.
      Assessment of chronic pain. I. Aspects of the reliability and validity of the visual analogue scale.
      ;
      • Kitaoka H.B.
      • Alexander I.J.
      • Adelaar R.S.
      • Nunley J.A.
      • Myerson M.S.
      • Sanders M.
      Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes.
      ;
      • Martin R.L.B.R.
      • Irrgang J.J.
      Development of the foot and ankle disability index (FADI).
      ;
      • Ware Jr., J.E.
      • Sherbourne C.D.
      The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection.
      ). In addition, baseline data from a case record form were gathered and a physical examination was performed.
      To assess the radiological findings on pre- and postoperative CT-scans with a higher diagnostic accuracy, the conventional CT-scans were converted into 3D CT-scans (
      • de Muinck Keizer R.O.
      • Meijer D.T.
      • van der Gronde B.A.
      • Teunis T.
      • Stufkens S.A.
      • Kerkhoffs G.M.
      • et al.
      Articular gap and step-off revisited: 3D quantification of operative reduction for posterior malleolar fragments.
      ;
      • Mangnus L.
      • Meijer D.T.
      • Stufkens S.A.
      • Mellema J.J.
      • Steller E.P.
      • Kerkhoffs G.M.
      • et al.
      Posterior malleolar fracture patterns.
      ;
      • Meijer D.T.
      • de Muinck Keizer R.J.
      • Doornberg J.N.
      • Sierevelt I.N.
      • Stufkens S.A.
      • Kerkhoffs G.M.
      • et al.
      Diagnostic accuracy of 2-dimensional computed tomography for articular involvement and fracture pattern of posterior malleolar fractures.
      ). The conventional CT-scans were uploaded in Vesalius3D (PS-Medtech, Amsterdam) and for each conventional CT-scan an STL-file with a three-dimensional image of the talocrural joint was created. Subsequently, the STL-file was imported in Blender (Blender Foundation, version 2.78c, Amsterdam). The tibial plafond was dissected from the talocrural joint in order to measure the articular surface areas of the tibia plafond and the PMF in cm2. The articular surface area included the tibial plafond, excluding the medial malleolus as the latter is not part of the weight-bearing area of the tibial plafond. Two researchers independently determined fragment size based on the 3D CT-scans, dividing the surface area of the PMF by the articular surface area of the tibial plafond. In addition, they determined comminution and Haraguchi type of the PMF, based on operation reports and on conventional and 3D CT-scans. In case of disagreement, a third researcher was consulted (
      • Haraguchi N.
      • Haruyama H.
      • Toga H.
      • Kato F.
      Pathoanatomy of posterior malleolar fractures of the ankle.
      ). A PMF was regarded as comminuted if it consisted of at least two smaller parts.
      Gait analysis was performed by one researcher using the protocol described in our previous studies (
      • van Hoeve S.
      • de Vos J.
      • Verbruggen J.P.
      • Willems P.
      • Meijer K.
      • Poeze M.
      Gait analysis and functional outcome after calcaneal fracture.
      ;
      • van Hoeve S.
      • de Vos J.
      • Weijers P.
      • Verbruggen J.
      • Willems P.
      • Poeze M.
      • et al.
      Repeatability of the Oxford foot model for kinematic gait analysis of the foot and ankle.
      ;
      • van Hoeve S.
      • Leenstra B.
      • Willems P.
      • Poeze M.
      • Meijer K.
      The effect of age and speed on foot and ankle kinematics assessed using a 4-segment foot model.
      ). One step (heel strike to toe-off) was divided into a loading and push-off phase.

      2.3 Statistics

      Data from the VICON system (NEXUS, UK) was translated with Matlab (MathWorks) into walking speed and RoM between hindfoot-tibia and forefoot-hindfoot in the sagittal, frontal and transverse planes (“flexion and extension”, “abduction and adduction”, and “inversion and eversion”, respectively) (
      • van Hoeve S.
      • de Vos J.
      • Weijers P.
      • Verbruggen J.
      • Willems P.
      • Poeze M.
      • et al.
      Repeatability of the Oxford foot model for kinematic gait analysis of the foot and ankle.
      ;
      • Wu G.
      • Siegler S.
      • Allard P.
      • Kirtley C.
      • Leardini A.
      • Rosenbaum D.
      • et al.
      ISB recommendation on definitions of joint coordinate system of various joints for the reporting of human joint motion--part I: ankle, hip, and spine.
      ). In SPSS (IBM Statistics), descriptive statistics, Fisher's exact test, Fisher-Freeman-Halton exact test, chi-squared test, independent samples t-test and one-way ANOVA were used to analyze baseline characteristics and gait kinematics. Correlations between gait kinematics, questionnaires, and CT-scans were assessed using the Pearson correlation coefficient. The validity of the different tests was evaluated by means of a method for comparing areas under the receiver operating characteristic (ROC) curves (AUCs), as previously described in detail (
      • Hanley J.A.
      • McNeil B.J.
      The meaning and use of the area under a receiver operating characteristic (ROC) curve.
      ;
      • Hanley J.A.
      • McNeil B.J.
      A method for comparing the areas under receiver operating characteristic curves derived from the same cases.
      ;
      • Nelson M.C.
      • Jensen N.K.
      The treatment of trimalleolar fractures of the ankle.
      ;
      • Wu G.
      • Siegler S.
      • Allard P.
      • Kirtley C.
      • Leardini A.
      • Rosenbaum D.
      • et al.
      ISB recommendation on definitions of joint coordinate system of various joints for the reporting of human joint motion--part I: ankle, hip, and spine.
      ). The optimal cut-off value for predicting excellent (AOFAS score > 90) functional outcome was calculated as the point with the greatest combined sensitivity and specificity. If a p-value <0.05 was found, the null hypothesis was rejected. The range of motion between the hindfoot and tibia during the push-off phase in the sagittal plane in healthy controls was estimated to be 12.5°(±4.0) (
      • van Hoeve S.
      • Houben M.
      • Vebruggen J.P.A.M.
      • Willems P.
      • Meijer K.
      • Poeze M.
      Gait analysis related to functional outcome in patients operated for ankle fractures.
      ). A clinically relevant decrease in range of motion after a trimalleolar fracture was estimated to be 4.5° for flexion-extension during both the loading and push-off phases (
      • van Hoeve S.
      • de Vos J.
      • Weijers P.
      • Verbruggen J.
      • Willems P.
      • Poeze M.
      • et al.
      Repeatability of the Oxford foot model for kinematic gait analysis of the foot and ankle.
      ). The number of patients to analyze was 24.

      3. Findings

      3.1 Patient characteristics

      The trimalleolar group consisted of 26 ft (Table 1). Eleven posterior malleolus fractures were Haraguchi type I, 11 were Haraguchi type II and 4 were Haraguchi type III. The mean (SD, min-max) size of the PMF as a percentage of the surface area of the posterior malleolus, based on conventional CT-scans and 3D-CT scans, was 18.5% (6.8, 7–33) and 18.8% (8.7, 6–40), respectively (Bland Altman: Beta −0.286, p = 0.1), indicating no proportional bias between the two methods. Fixation of the PMF was applied in 14 of the 26 patients; the method of fixation was decided by the surgeon, according to the AO Foot & Ankle Manual (2020); 30% and intra articular imvolvement or haraguchi type 2/ Bartonicek and rammel type 4.
      Table 1Baseline characteristics.
      Trimalleolar Ankle FracturesHealthy controlsp-value
      No. of subjects (no. of feet)26 (26)14 (26)
      Age (years)58.1 ± 18.6 (22–78)47.9 ± 16.6 (20–65)0.094
      Unpaired t-test.
      Gender (% of subjects)0.022
      p < 0.05.
      Fisher's exact test.
      Male10 (38)11 (79)
      Female16 (62)3 (21)
      Side (% of feet)0.577
      Fisher's exact test.
      Right16 (62)13 (50)
      Left10 (38)13 (50)
      Height (mm)1670.5 ± 112.5 (1535–1930)1785.3 ± 62.8 (1625–1850)<0.001
      p < 0.05.
      Unpaired t-test.
      Weight (kg)78.1 ± 15.1 (55–112)77.7 ± 8.8 (63–91)0.932
      Unpaired t-test.
      BMI (kg/m2)27.9 ± 4.4 (21.8–40.2)24.5 ± 3.1 (19.4–30.3)0.013
      p < 0.05.
      Unpaired t-test.
      Knee width (mm)100.4 ± 9.1 (80–120)103.2 ± 6.6 (92–114)0.204
      Unpaired t-test.
      Ankle width (mm)72.8 ± 6.1 (58–87)69.4 ± 4.9 (62–77)0.030
      p < 0.05.
      Unpaired t-test.
      Leg length (mm)860.0 ± 67.6 (760–990)926.1 ± 43.1 (790–970)<0.001
      p < 0.05.
      Unpaired t-test.
      Age, height, weight, BMI, knee width, ankle width and leg width are listed as means with standard deviation and range.
      a Unpaired t-test.
      b Fisher's exact test.
      low asterisk p < 0.05.
      Residual gap and step-off were within normal limits (<2 mm and < 1 mm, respectively) for all patients. Malposition of fixation occurred in 1 subject and removal of osteosynthesis material due to complaints was indicated in 7 patients. Revision surgery was not performed in any of the patients. Postoperatively, 22 patients received physical therapy, for a period ranging from 2 to 24 months. Mean time between surgery and gait analysis was 24 months (ranging from 6 to 52 months).
      The mean scores of the SF-36 physical functioning, FADI, VAS and AOFAS questionnaires were 70.4 ± 26.0, 81.3 ± 17.1, 1.6 ± 2.0, and 85.9 ± 12.7, respectively. The mean size of the PMF as a percentage of that of the posterior malleolus was 19% ± 6.8 based on conventional CT-scans, and 19% ± 8.7 based on 3D CT-scans (r2 = 0.66; p = 0.0001).
      Baseline characteristics were evaluated after subdividing the patients with trimalleolar fractures into three groups, with a PMF size ≤10%, 10–25% and ≥ 25% of the posterior malleolus, based on 3D CT-scans (Table 2). Fixation was applied in 0 of the 6 patients with a PMF size ≤10%, 9 of the 14 subjects with a PMF size 10–25% and 5 of the 6 patients with a PMF size ≥25% (p = 0.006). The time between surgery and the gait analysis differed significantly between patients with a fragment size >25% compared to those with a fragment size of 10–25%, although no correlation was found between time to gait analysis, the two gait parameters, and the functional outcome parameters.
      Table 2Baseline characteristics for posterior malleolus fragment size.
      Posterior malleolus fragment size ≤ 10%Posterior malleolus fragment size 10–25%Posterior malleolus fragment size ≥ 25%p-value
      No. of subjects6146
      Age (years)59.5 ± 21.2 (29–78)54.6 ± 17.9 (22–75)64.8 ± 19.0 (31–78)p = 0.536
      One-way ANOVA.
      Gender (% of subjects)p = 0.571
      Fisher-Freeman-Halton exact test.
      Male1 (17)6 (43)3 (50)
      Female5 (83)8 (57)3 (50)
      Side (% of feet)p = 0.647
      Fisher-Freeman-Halton exact test.
      Right3 (50)10 (71)3 (50)
      Left3 (50)4 (29)3 (50)
      Haraguchi classification (% of feet)p = 0.446
      Chi-squared test.
      Type I3 (50)5 (36)3 (50)
      Type II1 (17)7 (50)3 (50)
      Type III2 (33)2 (14)0 (0)
      ASA (% of subjects)p = 0.357
      Chi-squared test.
      Type I2 (33)4 (29)3 (50)
      Type II3 (50)10 (71)3 (50)
      Type III1 (17)0 (0)0 (0)
      Communition posterior malleolus fragmentp = 0.050
      Fisher-Freeman-Halton exact test.
      No4 (67)2 (14)3 (50)
      Yes2 (33)12 (86)3 (50)
      Fixation posterior malleolus fragment (% of subjects)0 (0)9 (64)5 (83)p = 0.006
      p < 0.05.
      Fisher-Freeman-Halton exact test.
      Removal osteosynthesis material (% of subjects)0 (0)4 (29)3 (50)p = 0.185
      Fisher-Freeman-Halton exact test.
      Malposition osteosynthesis material (% of subjects)1 (17)0 (0)0 (0)p = 0.462
      Fisher-Freeman-Halton exact test.
      Physical therapy (% of subjects)5 (83)11 (79)6 (100)p = 0.781
      Fisher-Freeman-Halton exact test.
      Days between surgery and gait analysis834 ± 441 (331–1571)549 ± 305 (203–1386)1088 ± 452 (290–1492)p = 0.021
      p < 0.05.
      One-way ANOVA.
      SF-36 physical functioning77.5 ± 25.1 (35–100)70.0 ± 24.8 (25–100)64.2 ± 32.5 (30−100)p = 0.690
      One-way ANOVA.
      FADI87.8 ± 13.6 (61.5–100)80.2 ± 15.7 (53.8–100)77.6 ± 23.7 (44.2–100)p = 0.561
      One-way ANOVA.
      VAS2.7 ± 3.0 (0–7)1.5 ± 1.7 (0–5)0.8 ± 1.2 (0–3)p = 0.257
      One-way ANOVA.
      AOFAS90.0 ± 12.2 (67–100)86.0 ± 11.0 (66–100)81.3 ± 17.3 (60–100)p = 0.517
      One-way ANOVA.
      Age, % posterior malleolus fragment size on 3D CT, days between surgery and gait analysis, SF-36 physical functioning, FADI, VAS and AOFAS are listed are listed as means with standard deviation and range.
      a Chi-squared test.
      b One-way ANOVA.
      c Fisher-Freeman-Halton exact test.
      low asterisk p < 0.05.
      Baseline characteristics and questionnaire scores did not differ significantly between the non-comminuted and comminuted groups or between the Haraguchi fracture types.

      3.2 Gait analysis

      The patients with the trimalleolar fractures had a significantly lower walking speed compared to the healthy controls (0.90 ± 0.26 vs. 1.20 ± 0.20; p < 0.001). The RoMs at normal walking speed of the patients with the trimalleolar fractures were comparable to the RoMs of the healthy controls at a slow walking speed (0.90 ± 0.26° vs. 0.88 ± 2.35°; p = 0.811) (Table 3) (
      • van Hoeve S.
      • Leenstra B.
      • Willems P.
      • Poeze M.
      • Meijer K.
      The effect of age and speed on foot and ankle kinematics assessed using a 4-segment foot model.
      ).
      Table 3Speed and gait analysis in patients with trimalleolar ankle fractures and healthy control patients.
      Trimalleolar Ankle FracturesPosterior malleolus fragment ≤ 10%Posterior malleolus fragment 10–25%Posterior malleolus fragment ≥ 25%Healthy Controlsp-value a
      Speed0.90 ± 0.26 (normal)0.86 ± 0.300.93 ± 0.260.84 ± 0.251.20 ± 0.20 (normal)

      0.88 ± 0.21 (slow)
      <0.001
      p-value<0.05.


      0.811
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.725
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Forefoot-hindfoot
      Flexion/extension loading phase8.85 ± 2.5911.41 ± 1.938.15 ± 2.487.94 ± 1.918.04 ± 2.350.242
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.015
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Abduction/adduction loading phase3.50 ± 1.463.87 ± 1.943.23 ± 1.063.74 ± 1.904.78 ± 1.470.003
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.616
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Inversion/eversion loading phase6.14 ± 2.496.48 ± 3.006.11 ± 2.535.86 ± 2.276.75 ± 2.020.331
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.914
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Flexion/extension push-off phase16.34 ± 5.3717.76 ± 4.4916.83 ± 5.9713.77 ± 4.5215.52 ± 3.720.527
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.401
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Abduction/adduction push-off phase8.14 ± 3.078.46 ± 2.838.40 ± 3.287.24 ± 3.158.86 ± 2.860.387
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.728
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Inversion/eversion push-off phase8.70 ± 2.258.39 ± 2.268.54 ± 2.269.37 ± 2.508.61 ± 1.880.879
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.716
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Hindfoot-tibia
      Flexion/extension loading phase6.59 ± 2.317.43 ± 2.606.44 ± 2.236.07 ± 2.439.65 ± 3.04<0.001
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.585
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Abduction/adduction loading phase11.03 ± 5.1011.12 ± 3.8711.13 ± 5.5910.73 ± 5.8111.06 ± 3.640.979
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.987
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Inversion/eversion loading phase3.99 ± 2.054.16 ± 1.954.20 ± 2.273.34 ± 1.756.15 ± 2.03<0.001
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.694
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Flexion/extension push-off phase8.34 ± 2.828.93 ± 3.057.99 ± 2.418.56 ± 3.8011.72 ± 4.140.001
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.787
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Abduction/adduction push-off phase13.39 ± 6.3214.18 ± 5.8312.36 ± 4.8114.98 ± 9.9310.30 ± 4.140.042
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.674
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Inversion/eversion push-off phase5.81 ± 2.985.51 ± 3.075.85 ± 2.476.03 ± 4.379.40 ± 4.040.001
      p-value<0.05.
      Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      0.958
      Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      Results are listed as means in degrees with standard deviation and range.
      a Unpaired t-test comparing trimalleolar ankle fractures with healthy controls.
      b Unpaired t-test comparing trimalleolar ankle fractures having a posterior malleolus fragment ≤10% with 10–25% and ≥ 25% of the total articular surface.
      low asterisk p-value<0.05.
      The RoMs between the hindfoot and tibia in the sagittal and transverse planes in the loading phase in the patients with trimalleolar fractures were significantly lower than those in the healthy controls (6.59° ± 2.3° vs. 9.65° ± 3.05°; p < 0.001 and 3.99° ± 2.05° vs. 6.15° ± 2.03°; p < 0.001) (Table 3). Similarly, the RoMs between the hindfoot and tibia in the sagittal (8.34° ± 2.82° vs. 11.72° ± 4.14°; p = 0.001), frontal (13.39° ± 6.32° vs. 10.30° ± 4.14°; p = 0.042) and transverse planes (5.81° ± 2.98° vs. 9.40° ± 4.04°; p = 0.001) in the push-off phase differed significantly between the patients with trimalleolar fractures and the healthy controls. The RoM between the forefoot and hindfoot in the frontal plane in the loading phase was significantly lower in the patients with trimalleolar fractures compared to the healthy controls (3.50 ± 1.46 vs. 4.78 ± 1.47; p = 0.003).
      The RoM between the hindfoot and tibia in the sagittal plane in the push-off phase correlated significantly with the SF-36 physical functioning, FADI, VAS and AOFAS scores (r2 = 0.402; p = 0.003, r2 = 0.404; p = 0.003, r2 = −0.348; p = 0.012, and r2 = 0.358; p = 0.009, respectively) (Table 4). In addition, this RoM correlated significantly with the size of the PMF as a percentage of the posterior malleolus based on conventional CT-scans and the absolute and relative surface area of the PMF on 3D CT-scans (r2 = −0.290; p = 0.037, r2 = −0.352; p = 0.010, and r2 = −0.382; p = 0.005, respectively). The size of the PMF as a percentage of the tibial plafond, based on conventional and 3D CT-scans, correlated significantly with the SF-36 physical functioning, FADI, VAS and AOFAS scores (r2 = −0.461; p = 0.001, r2 = −0.442; p = 0.001, r2 = 0.283; p = 0.042, and r2 = −0.433; p = 0.001, respectively, for conventional CT-scans and r2 = −0.496; p < 0.001, r2 = −0.462; p = 0.001, r2 = 0.300; p = 0.031, and r2 = −0.467; p < 0.001, respectively, for 3D CT-scans). The optimal cut-off value for the size of the PMF as a percentage of the posterior malleolus to predict an excellent (≥90) AOFAS score was <15%, as determined in the ROC curve analysis as the point with greatest combined sensitivity and specificity.
      Table 4Correlation between ROMs, PROMs and size of the posterior malleolus fragment.
      SF-36 PHFFADIVASAOFASSurface area posterior malleolus (conventional CT)Percentage posterior malleolus (conventional CT)Surface area posterior malleolus (3D CT)Percentage posterior malleolus (3D CT)ComminutionHaraguchi type
      ROM hindfoot-tibia
      Flexion/extension loading phase0.2340.259−0.2120.227−0.368
      p < 0.01.
      −0.438
      p < 0.01.
      −0.438
      p < 0.01.
      −0.466
      p < 0.01.
      −0.2290.226
      Flexion/extension push-off phase0.402

      p < 0.01.
      0.404
      p < 0.01.
      −0.348
      p < 0.05.
      0.358
      p < 0.01.
      −0.228−0.290
      p < 0.05.
      −0.352
      p < 0.05.
      −0.382
      p < 0.01.
      0.2040.037
      ROM forefoot-hindfoot
      Flexion/ extension loading phase0.0120.0330.031−0.089−0.543
      p < 0.01.
      −0.615
      p < 0.01.
      −0.529
      p < 0.01.
      −0.525
      p < 0.01.
      −0.3240.208
      Flexion/ extension push-off phase0.423
      p < 0.05.
      0.380−0.3130.231−0.142−0.218−0.155−0.2900.3040.171
      Radiographic findings
      Surface area posterior malleolus fragment (conventional CT)−0.383
      p < 0.01.
      −0.367
      p < 0.01.
      0.217−0.365
      p < 0.01.
      10.984
      p < 0.01.
      0.903
      p < 0.01.
      0.877
      p < 0.01.
      0.290−0.012
      Percentage posterior malleolus fragment (conventional CT)−0.461
      p < 0.01.
      −0.442
      p < 0.01.
      0.283
      p < 0.05.
      −0.433
      p < 0.01.
      0.984
      p < 0.01.
      10.913
      p < 0.01.
      0.912
      p < 0.01.
      0.349−0.115
      Surface area posterior malleolus fragment (3D CT)−0.421
      p < 0.01.
      −0.405
      p < 0.01.
      0.259−0.418
      p < 0.01.
      0.903
      p < 0.01.
      0.913
      p < 0.01.
      10.980
      p < 0.01.
      0.024−0.072
      Percentage posterior malleolus fragment (3D CT)−0.496
      p < 0.01.
      −0.462
      p < 0.01.
      0.300
      p < 0.05.
      −0.467
      p < 0.01.
      0.877
      p < 0.01.
      0.912
      p < 0.01.
      0.980
      p < 0.01.
      1−0.067−0.221
      Comminution groups0.3120.187−0.3180.2630.2900.3490.024−0.06710.066
      Haraguchi type groups−0.100−0.2490.110−0.321−0.01−0.115−0.072−0.2210.0661
      Results are listed as Pearson correlation coefficient (r).
      low asterisk p < 0.05.
      low asterisklow asterisk p < 0.01.
      Patients with a PMF size ≤10% had a significantly greater RoM between the forefoot and hindfoot in the sagittal plane than patients with a PMF size 10–25% and ≥ 25% (11.41 ± 1.93 vs. 8.15 ± 2.48 vs. 7.94 ± 1.91; p = 0.015) (Fig 1).
      Fig. 1
      Fig. 1Tibia-hindfoot and hindfoot-forefoot flexion and extension during gait in patients with posterior malleolar fractures and healthy controls.
      The RoM in the sagittal plane and the PROMs were analyzed for correlations within the PMF subgroups for comminution and size (Table 4). The RoM between the forefoot and hindfoot in the sagittal plane in the loading phase correlated significantly with the PMF size groups (r2 = −0.464; p = 0.017). The RoM between the forefoot and hindfoot in the sagittal plane in the push-off phase correlated significantly with the SF-36 physical functioning (r2 = 0.423; p = 0.031).
      Patients with a non-comminuted PMF had significantly lower RoM between the forefoot and hindfoot in the transverse plane in the loading phase and in the frontal plane in the push-off phase (4.77 ± 1.90 vs. 6.86 ± 2.51; p = 0.039 and 6.68 ± 1.88 vs. 8.92 ± 3.33; p = 0.039) (Table 4).
      Other ROMs between hindfoot and tibia were not significantly different between groups.

      4. Interpretation

      Our findings show that patients with posterior malleolar fractures as part of trimalleolar ankle fractures had abnormal gait kinematics. Patients with a PMF had less flexion-extension in the talocrural joint and a lower walking speed. The ROMs at normal walking speed of the patients with the trimalleolar fractures were comparable to the ROMs of the healthy controls at a slow walking speed (0.90 ± 0.26° vs. 0.88 ± 2.35°; p = 0.811), as walking speed influences the ankle and foot kinematics. (
      • van Hoeve S.
      • Leenstra B.
      • Willems P.
      • Poeze M.
      • Meijer K.
      The effect of age and speed on foot and ankle kinematics assessed using a 4-segment foot model.
      ) The gait kinematics of the patients with a PMF correlated with the patient-reported outcome measures (PROM). In addition, the gait kinematics correlated with radiographic findings. PMF size as determined by 3D-CT surface area measurement showed a significant correlation with the ROMs between both the talo-crural and the midfoot joints, while comminution and fracture configuration, using the Haraguchi classification, showed no relationship with functional outcome.
      So far, no studies have been published that evaluated the influence of size, comminution and fracture configuration of the PMF on gait kinematics in trimalleolar ankle fractures. To date, only few published studies have used a multi-segment foot model to compare the ankle kinematics of patients with ankle fractures with those of healthy controls. Wang et al. performed a gait analysis, using a modified OFM, in 18 patients with surgically treated ankle fractures one year postoperatively, and compared them with the non-injured side and with healthy controls (
      • Wang R.
      • Thur C.K.
      • Gutierrez-Farewik E.M.
      • Wretenberg P.
      • Brostrom E.
      One year follow-up after operative ankle fractures: a prospective gait analysis study with a multi-segment foot model.
      ). Six of the 18 ankle fracture patients included had a trimalleolar ankle fracture. In line with our findings, the fractured side showed lower RoM and plantarflexion in the sagittal plane between the hindfoot and tibia and lower RoM between the forefoot and hindfoot, compared with the uninjured side and healthy controls. Segal et al. examined 41 ankle fracture patients, using a computerized mat (
      • Segal G.
      • Elbaz A.
      • Parsi A.
      • Heller Z.
      • Palmanovich E.
      • Nyska M.
      • et al.
      Clinical outcomes following ankle fracture: a cross-sectional observational study.
      ). The patients with trimalleolar fractures showed significantly lower walking speed, and step length, single limb support asymmetry, and lower SF-36 and AOFAS scores compared to healthy controls. Finally, van Hoeve et al. found similar ROMs between hindfoot and tibia as the present study and reported significantly lower RoM between hindfoot and tibia in the sagittal plane in patients with ankle fractures compared to the healthy controls (
      • van Hoeve S.
      • Houben M.
      • Vebruggen J.P.A.M.
      • Willems P.
      • Meijer K.
      • Poeze M.
      Gait analysis related to functional outcome in patients operated for ankle fractures.
      ). In addition, the RoM of the talocrural joint in flexion and extension was significantly lower in patients with a trimalleolar fracture (n = 12) compared to those with a unimalleolar fracture (n = 10), but not those with a bimalleolar fracture (n = 11). The functional outcome, defined by FADI and AOFAS scores, decreased with an increasing number of fractured malleoli.
      Within our group, the RoM in the sagittal plane at the hindfoot and tibia and between forefoot and hindfoot also correlated significantly with the fragment size of the posterior malleolus. A larger involvement of the articular area of the tibial plafond was associated with a lower RoM in the talocrural joint as well as the joints between the hindfoot and the forefoot. The exact cut-off point for a posterior malleolus fracture to be significant remains controversial. Since the case series by Nelson and Jenson in 1940, it became accepted that a posterior malleolus in a trimalleolar fracture should be internally fixated when the size of the fragment exceeds 33% of the joint surface or, according to several other authors, if the PMF exceeds 25% (
      • McLaughlin H.L.
      Trauma.
      ;
      • Nelson M.C.
      • Jensen N.K.
      The treatment of trimalleolar fractures of the ankle.
      ). Other authors found no differences in outcome between medium-size (5–25%) and large (>25%) PMFs (
      • Court-Brown C.M.
      • Caesar B.
      Epidemiology of adult fractures: a review.
      ). Based upon these data, restoration of the anatomical alignment of PMF sizes between 5% and 25% may lead to improved outcome, and the same goes for fragments larger than 25% (
      • Broos P.L.
      • Bisschop A.P.
      Operative treatment of ankle fractures in adults: correlation between types of fracture and final results.
      ;
      • Jaskulka R.A.
      • Ittner G.
      • Schedl R.
      Fractures of the posterior tibial margin: their role in the prognosis of malleolar fractures.
      ;
      • Lindsjo U.
      Operative treatment of ankle fracture-dislocations. A follow-up study of 306/321 consecutive cases.
      ). In our study we did not find that a specific cut-off point can be used to determine the indication for surgical intervention. Notably, however, all patients had a reconstruction of the tibial plafond with good alignment, congruence and anatomical reduction on the X-ray at day one and at 6 weeks postoperative, and also on the CT direct postoperative (gap <2 mm and step-off <1 mm). Other authors have emphasized the importance of anatomic reconstruction over the exact cut-off point for fixation (
      • Evers J.
      • Barz L.
      • Wahnert D.
      • Gruneweller N.
      • Raschke M.J.
      • Ochman S.
      Size matters: the influence of the posterior fragment on patient outcomes in trimalleolar ankle fractures.
      ). This indicates that it is not only the presence of the trimalleolar fracture complex which has significant impact on the gait parameters, but also the size of the posterior malleolus fragment. The present study was, however, not designed to answer the question whether correction of smaller PMFs improves functional outcome, while postoperative residual abnormalities were within normal range.
      Additional reason for the different findings is that not only de size and shape but also location of the fragment or the integrity of the fibular notch is responsible for functional and patient reported outcome, and the Haraguchi classification would not be sufficient enough to identify. Bartoníček, J et al. reported a CT based classification and considers the size, shape and location of the fragment, stability of the tibio-talar joint and the integrity of the fibular notch (
      • Bartoníček J.
      • et al.
      • Rammelt S.
      • Kostlivý K.
      • et al.
      Anatomy and classification of the posterior tibial fragment in ankle fractures.
      ). Interestingly, the size of the PMF also correlated with the RoM in the midfoot, joints not primarily affected by the fractures. In these patients a higher flexion-extension RoM correlated with smaller sizes of the PMF, while patients with a small PMF (<10%) had a significantly higher RoM in the midfoot than the healthy controls. Obviously, these data indicate that the midfoot compensates for the loss of motion in PMF complexes of smaller sizes, but perhaps this compensation fails with larger size PMFs. Such a compensation after trauma within the foot kinematics of the injured ankle has also been observed by others (
      • Stevens J.
      • Meijer K.
      • Bijnens W.
      • Fuchs M.C.
      • van Rhijn L.W.
      • Hermus J.P.
      • van Hoeve S.
      • Poeze M.
      • Witlox A.M.
      Gait analysis of foot compensation after arthrodesis of the first metatarsophalangeal joint.
      ).
      A number of remarks should be made when interpreting the results of this study. The size of the PMF as a percentage of the posterior malleolus based on conventional and 3D CT-scans correlated well with each other, unlike what was found in a previous study by Meijer et al. (
      • Meijer D.T.
      • de Muinck Keizer R.J.
      • Doornberg J.N.
      • Sierevelt I.N.
      • Stufkens S.A.
      • Kerkhoffs G.M.
      • et al.
      Diagnostic accuracy of 2-dimensional computed tomography for articular involvement and fracture pattern of posterior malleolar fractures.
      ) However, the absolute and relative surface areas of the PMF on 3D CT-scans correlated better with the RoM in the sagittal plane between hindfoot and tibia than the absolute and relative surface areas of the PMF on conventional CT-scans, justifying the use of 3D CT-scans in addition to the conventional CT analysis. In addition, the OFM measures solely the ROMs between the tibia and hindfoot. It does not measure the kinematics of the talocrural and subtalar joints separately (
      • van Hoeve S.
      • de Vos J.
      • Verbruggen J.P.
      • Willems P.
      • Meijer K.
      • Poeze M.
      Gait analysis and functional outcome after calcaneal fracture.
      ). For example, there was a significant difference in RoM between the hindfoot and tibia in the transverse plane in both the loading and push-off phase between the patients with trimalleolar ankle fractures and the healthy controls, which indicates that a trimalleolar fracture not only affects the talocrural joint kinematics, but also those of the subtalar joint. Both of these are located between the tibia and hindfoot and have movements in the sagittal and transverse planes (
      • Brockett C.L.
      • Chapman G.J.
      Biomechanics of the ankle.
      ). Also no large group of patients were analyzed with 26 patients in the fracture group. Finally, the time between surgery and gait analysis of our patients varied between 6 and 52 months (
      • Beckenkamp P.R.
      • Lin C.W.
      • Chagpar S.
      • Herbert R.D.
      • van der Ploeg H.P.
      • Moseley A.M.
      Prognosis of physical function following ankle fracture: a systematic review with meta-analysis.
      ). Additional improvement after gait analysis cannot be ruled out in some patients, although no correlation was found between time to gait analysis and functional outcome.

      5. Conclusion

      In conclusion, this study shows that patients with trimalleolar ankle fractures have a lower walking speed and reduced RoM in the talocrural joint compared to healthy controls. In addition, the lower RoM in the talocrural joint is associated with poorer functional outcome parameters. Comminution and Haraguchi posterior malleolus fracture classification had no influence on functional outcome, whereas PMF size showed a significant correlation with the RoM of the foot and ankle and the functional outcome.
      Declaration of Competing Interest
      None.

      References

        • Agostini V.
        • Ganio D.
        • Facchin K.
        • Cane L.
        • Moreira Carneiro S.
        • Knaflitz M.
        Gait parameters and muscle activation patterns at 3, 6 and 12 months after total hip arthroplasty.
        J. Arthroplast. 2014; 29: 1265-1272
        • Bartoníček J.
        • et al.
        • Rammelt S.
        • Kostlivý K.
        • et al.
        Anatomy and classification of the posterior tibial fragment in ankle fractures.
        Arch. Orthop. Trauma Surg. 2015; 135: 505-516https://doi.org/10.1007/s00402-015-2171-4
        • Beckenkamp P.R.
        • Lin C.W.
        • Chagpar S.
        • Herbert R.D.
        • van der Ploeg H.P.
        • Moseley A.M.
        Prognosis of physical function following ankle fracture: a systematic review with meta-analysis.
        J. Orthop. Sports Phys. Ther. 2014; 44 (841–51, b2)
        • Becker H.P.
        • Rosenbaum D.
        • Kriese T.
        • Gerngross H.
        • Claes L.
        Gait asymmetry following successful surgical treatment of ankle fractures in young adults.
        Clin. Orthop. Relat. Res. 1995; 311: 262-269
        • Belcher G.L.
        • Radomisli T.E.
        • Abate J.A.
        • Stabile L.A.
        • Trafton P.G.
        Functional outcome analysis of operatively treated malleolar fractures.
        J. Orthop. Trauma. 1997; 11: 106-109
        • Blom R.P.
        • Hayat B.
        • Al-Dirini R.M.A.
        • Sierevelt I.
        • Kerkhoffs G.M.M.J.
        • Goslings J.C.
        • Jaarsma R.L.
        • Doornberg J.N.
        • EF3X-trial Study Group
        Posterior malleolar ankle fractures.
        Bone Joint J. 2020; 102-B (Sep. (PMID: 32862684)): 1229-1241https://doi.org/10.1302/0301-620X.102B9.BJJ-2019-1660.R1
        • Brockett C.L.
        • Chapman G.J.
        Biomechanics of the ankle.
        Orthop. Trauma. 2016; 30: 232-238
        • Broos P.L.
        • Bisschop A.P.
        Operative treatment of ankle fractures in adults: correlation between types of fracture and final results.
        Injury. 1991; 22: 403-406
        • Carlsson A.M.
        Assessment of chronic pain. I. Aspects of the reliability and validity of the visual analogue scale.
        Pain. 1983; 16: 87-101
        • Court-Brown C.M.
        • Caesar B.
        Epidemiology of adult fractures: a review.
        Injury. 2006; 37: 691-697
        • de Muinck Keizer R.O.
        • Meijer D.T.
        • van der Gronde B.A.
        • Teunis T.
        • Stufkens S.A.
        • Kerkhoffs G.M.
        • et al.
        Articular gap and step-off revisited: 3D quantification of operative reduction for posterior malleolar fragments.
        J. Orthop. Trauma. 2016; 30: 670-675
        • De Vries J.S.
        • Wijgman A.J.
        • Sierevelt I.N.
        • Schaap G.R.
        Long-term results of ankle fractures with a posterior malleolar fragment.
        J. Foot Ankle Surg. 2005; 44: 211-217
        • Donken C.C.
        • Al-Khateeb H.
        • Verhofstad M.H.
        • van Laarhoven C.J.
        Surgical versus conservative interventions for treating ankle fractures in adults.
        Cochrane Database Syst. Rev. 2012; (Cd008470)
        • Drijfhout van Hooff C.C.
        • Verhage S.M.
        • Hoogendoorn J.M.
        Influence of fragment size and postoperative joint congruency on long-term outcome of posterior malleolar fractures.
        Foot Ankle Int. 2015; 36: 673-678
        • Evers J.
        • Barz L.
        • Wahnert D.
        • Gruneweller N.
        • Raschke M.J.
        • Ochman S.
        Size matters: the influence of the posterior fragment on patient outcomes in trimalleolar ankle fractures.
        Injury. 2015; 46: S109-S113
        • Hanley J.A.
        • McNeil B.J.
        The meaning and use of the area under a receiver operating characteristic (ROC) curve.
        Radiology. 1982; 143: 29-36
        • Hanley J.A.
        • McNeil B.J.
        A method for comparing the areas under receiver operating characteristic curves derived from the same cases.
        Radiology. 1983; 148: 839-843
        • Haraguchi N.
        • Haruyama H.
        • Toga H.
        • Kato F.
        Pathoanatomy of posterior malleolar fractures of the ankle.
        J. Bone Joint Surg. Am. 2006; 88: 1085-1092
        • Jaskulka R.A.
        • Ittner G.
        • Schedl R.
        Fractures of the posterior tibial margin: their role in the prognosis of malleolar fractures.
        J. Trauma. 1989; 29: 1565-1570
        • Kitaoka H.B.
        • Alexander I.J.
        • Adelaar R.S.
        • Nunley J.A.
        • Myerson M.S.
        • Sanders M.
        Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes.
        Foot Ankle Int. 1994; 15: 349-353
        • Lindsjo U.
        Operative treatment of ankle fracture-dislocations. A follow-up study of 306/321 consecutive cases.
        Clin. Orthop. Relat. Res. 1985; 199: 28-38
        • Losch A.
        • Meybohm P.
        • Schmalz T.
        • Fuchs M.
        • Vamvukakis F.
        • Dresing K.
        • et al.
        Functional results of dynamic gait analysis after 1 year of hobby-athletes with a surgically treated ankle fracture.
        Sportverletzung Sportschaden. 2002; 16: 101-107
        • Mangnus L.
        • Meijer D.T.
        • Stufkens S.A.
        • Mellema J.J.
        • Steller E.P.
        • Kerkhoffs G.M.
        • et al.
        Posterior malleolar fracture patterns.
        J. Orthop. Trauma. 2015; 29: 428-435
        • Martin R.L.B.R.
        • Irrgang J.J.
        Development of the foot and ankle disability index (FADI).
        J. Orthop. Sports Phys. Ther. 1999; 29 (A32–A3)
        • McLaughlin H.L.
        Trauma.
        W.B. Saunders Co., Philadelphia1959: 357-360
        • Meijer D.T.
        • Doornberg J.N.
        • Sierevelt I.N.
        • Mallee W.H.
        • van Dijk C.N.
        • Kerkhoffs G.M.
        • et al.
        Guesstimation of posterior malleolar fractures on lateral plain radiographs.
        Injury. 2015; 46: 2024-2029
        • Meijer D.T.
        • de Muinck Keizer R.J.
        • Doornberg J.N.
        • Sierevelt I.N.
        • Stufkens S.A.
        • Kerkhoffs G.M.
        • et al.
        Diagnostic accuracy of 2-dimensional computed tomography for articular involvement and fracture pattern of posterior malleolar fractures.
        Foot Ankle Int. 2016; 37: 75-82
        • Nantel J.
        • Termoz N.
        • Vendittoli P.A.
        • Lavigne M.
        • Prince F.
        Gait patterns after total hip arthroplasty and surface replacement arthroplasty.
        Arch. Phys. Med. Rehabil. 2009; 90: 463-469
        • Nelson M.C.
        • Jensen N.K.
        The treatment of trimalleolar fractures of the ankle.
        Surg. Gynaecol. Obst. 1940; 71: 509-514
        • Odak S.
        • Ahluwalia R.
        • Unnikrishnan P.
        • Hennessy M.
        • Platt S.
        Management of Posterior Malleolar Fractures: a systematic review.
        J. Foot Ankle Surg. 2016; 55: 140-145
        • Segal G.
        • Elbaz A.
        • Parsi A.
        • Heller Z.
        • Palmanovich E.
        • Nyska M.
        • et al.
        Clinical outcomes following ankle fracture: a cross-sectional observational study.
        J. Foot Ankle Res. 2014; 7: 50
        • Stevens J.
        • Meijer K.
        • Bijnens W.
        • Fuchs M.C.
        • van Rhijn L.W.
        • Hermus J.P.
        • van Hoeve S.
        • Poeze M.
        • Witlox A.M.
        Gait analysis of foot compensation after arthrodesis of the first metatarsophalangeal joint.
        Foot Ankle Int. 2017; 38 (Feb): 181-191
        • Tejwani N.C.
        • Pahk B.
        • Egol K.A.
        Effect of posterior malleolus fracture on outcome after unstable ankle fracture.
        J. Trauma. 2010; 69: 666-669
        • Tenenbaum S.
        • Shazar N.
        • Bruck N.
        • Bariteau J.
        Posterior malleolus fractures.
        Orthop. Clin. North Am. 2017; 48: 81-89
        • van den Bekerom M.P.
        • Haverkamp D.
        • Kloen P.
        Biomechanical and clinical evaluation of posterior malleolar fractures. A systematic review of the literature.
        J. Trauma. 2009; 66: 279-284
        • van Hoeve S.
        • de Vos J.
        • Verbruggen J.P.
        • Willems P.
        • Meijer K.
        • Poeze M.
        Gait analysis and functional outcome after calcaneal fracture.
        J. Bone Joint Surg. Am. 2015; 97: 1879-1888
        • van Hoeve S.
        • de Vos J.
        • Weijers P.
        • Verbruggen J.
        • Willems P.
        • Poeze M.
        • et al.
        Repeatability of the Oxford foot model for kinematic gait analysis of the foot and ankle.
        Clin. Res. Foot Ankle. 2015; 3
        • van Hoeve S.
        • Leenstra B.
        • Willems P.
        • Poeze M.
        • Meijer K.
        The effect of age and speed on foot and ankle kinematics assessed using a 4-segment foot model.
        Medicine (Baltimore). 2017; 96e7907
        • van Hoeve S.
        • Houben M.
        • Vebruggen J.P.A.M.
        • Willems P.
        • Meijer K.
        • Poeze M.
        Gait analysis related to functional outcome in patients operated for ankle fractures.
        J. Orthop. Res. 2018; (Jun 19)https://doi.org/10.1002/jor.24071
        • Veltman E.S.
        • Halma J.J.
        • de Gast A.
        Longterm outcome of 886 posterior malleolar fractures: a systematic review of the literature.
        Foot Ankle Surg. 2016; 22: 73-77
        • Wang R.
        • Thur C.K.
        • Gutierrez-Farewik E.M.
        • Wretenberg P.
        • Brostrom E.
        One year follow-up after operative ankle fractures: a prospective gait analysis study with a multi-segment foot model.
        Gait Posture. 2010; 31: 234-240
        • Ware Jr., J.E.
        • Sherbourne C.D.
        The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection.
        Med. Care. 1992; 30: 473-483
        • Wright C.J.
        • Arnold B.L.
        • Coffey T.G.
        • Pidcoe P.E.
        Repeatability of the modified Oxford foot model during gait in healthy adults.
        Gait Posture. 2011; 33: 108-112
        • Wu G.
        • Siegler S.
        • Allard P.
        • Kirtley C.
        • Leardini A.
        • Rosenbaum D.
        • et al.
        ISB recommendation on definitions of joint coordinate system of various joints for the reporting of human joint motion--part I: ankle, hip, and spine.
        Int. Soc. Biomech. J. Biomech. 2002; 35: 543-548
        • Xu H.L.
        • Li X.
        • Zhang D.Y.
        • Fu Z.G.
        • Wang T.B.
        • Zhang P.X.
        • et al.
        A retrospective study of posterior malleolus fractures.
        Int. Orthop. 2012; 36: 1929-1936