A Systematic Review of the Routine Removal of Syndesmosis Screws in Traumatic Ankle Fractures

Shaunak S; AhmedN; Singh S

Review Article

Austin J Orthopade & Rheumatol. 2022; 9(2): 1113.

A Systematic Review of the Routine Removal of Syndesmosis Screws in Traumatic Ankle Fractures

Shaunak S¹*, AhmedN² and Singh S²

1Department of Trauma & Orthopaedics, Medway NHS Foundation Trust, United Kingdom

2Hichingbroke Hospital, Parkway Hinchingbrooke, Huntingdon PE29 6NT, United Kingdom

*Corresponding author: Shalin Shaunak, Department of Trauma & Orthopaedics, Medway NHS Foundation Trust, Windmill Road, Gillingham, Kent, ME7 5NY, United Kingdom

Received: August 08, 2022; Accepted: September 01, 2022; Published: September 08, 2022

Introduction

The ankle syndesmosis is a critical structure, conferring a great degree of stability to the ankle mortise comprising part of a complex framework of ligaments responsible for 90% of the resistance to lateral displacement of the fibula [1]. The components of the syndesmosis are the Anterior-Inferior Tibiofibular Ligament (AITFL), Posterior- Inferior Tibiofibular Ligament (PITFL), Transverse Tibiofibular Ligament (TTFL) and Interosseous Ligament (IOL).

With approximately one in seven ankle fractures associated with a syndesmotic injury it is a common pattern [2]. The injury itself is most often found intra-operatively after stable fixation of malleolar fractures, indicated by persistent instability. This instability is often dealt with through the use of a bridging syndesmotic screw which restores the ankle mortise, confers stability but leaves the original syndesmotic injury relatively untouched.

Table 1:




  
    Author & Year Published
    Study type
    OLOE
    Screws Removed
    Screws Retained
    Mean F/U time (mths)
    Mean outcome
    Conclusion
  
  
    Bell et al 2006
    Prospective Cohort    Study
    4
    Removal 6-12/52

      N = 23
    Retention, N = 7

      Broken, N = 2

      Intact N = 5
    15
    Baird and Jackson    score 88 vs 86 (p =0.79)
    No statistically    significant difference
  
  
    Boyle et al 2014
    Randomised Control    Trial
    2b
    Removal >12/52

      N = 26
    Retention, N= 25

      Broken, N = 9

      Intact, N = 16
    12
    OMAS 86.7 vs 82.4 (p    = 0.367)
    No statistically    significant difference
  
  
    Egol et al 2010
    Retrospective Cohort    Study
    4
    Removal >12/52

      N = 11
    Retention, N= 68

      Broken, N = 15

      Intact, N = 53
    12
    AOFAS NP (p = 0.82)
    No statistically    significant difference
  
  
    Francesco et al 2019
    Retrospective Cohort Study 
    4
    Removal 6-8/52

      N = 65
    Retention, N = 25

      Broken, N = 8

      Intact, N= 17
    12
    AOFAS

      94 vs 99 (P = 0.56)
      OMAS

        95 vs 92.5 (p =    0.081)
    No statistically    significant difference
  
  
    Heck et al. 2017
    Retrospective Cohort Study
    4
    Removal 12/52

      N = 11
    Retention, N = 24
    6
    AOFAS

      90.3 vs 87.7 (p =    0.340)
    No statistically    significant difference
  
  
    Hamid et al. 2009
    Cross-sectional    Cohort Study
    4
    Removal 11-20/52
    Retention, N= 37

      Broken, N = 10

      Intact, N = 27
    30
    AOFAS 85.8 vs 85.59    (p 0.96)
    No statistically    significant difference
  
  
    Hoiness et al 2004
    Randomised Control    Trial
    2b
    Removal 9.5/52

      N = 30
    Retention, N= 34

      Broken, N = 3

      Intact, N = 31
    12
    OMAS 83.3 vs 88.8 (p=    0.192)
    No statistically    significant difference
  
  
    Kaftandziev et al    2015
    Cross Sectional    Cohort Study
    4
    Removal 8-12/52

      N = 23
    Retention, N= 59

      Broken, N = 13

      Intact, N = 46
    12
    AOFAS 83 vs 88.03 (p    =0.043)
    Patients with    retained screws had PROMs
  
  
    Kim et al

      2016
    Randomised Control    Trial
    2b
    Removal 3/12

      N = 56
    Retention, N = 56

      Broken, N = 8

      Intact, N = 48
    12
    AOFAS

      75.1 vs 77.07 (p =    0.487)
    No statistically    significant difference
  
  
    Kolodziej et al 2010
    Cross Sectional    Cohort Study
    4
    Removal 7-22/52

      N = 13
    Retention, N= 20

      Broken, N =8 15

      Intact, N = 12
    37
    AOFAS 89 vs 86.2
    No statistically    significant difference
  
  
    Manjoo et al 2010
    Cross Sectional    Cohort Study
    4
    Removal >40/52

      N = 12
    Retention, N= 64

      Broken, N = 44

      Intact, N = 20
    23
    OMAS 66.80 vs 57.29    (p < 0.001)
    Patients with intact    screws have worse outcomes than those with removed or broken screws
  
  
    Miller et al. 2010
    Prospective cohort    study
    4
    Removal 12/52

      N = 42
    Retention, N = 42    (all removed after assessment with screws in situ)
    13
    OMAS NP but suggested    significant improvement after removal (p = 0.003)
    Significant    improvement after removal of screws
  
  
    Moon et al. 2020
    Retrospective cohort    study
    4
    Removal 12/52

      N = 28
    Retention, N = 28

      Broken, N = 1

      Intact, N = 27
    >24
    AOFAS

      75.1 vs 77.07 (0.487)
    No statistically    significant difference
  
  
    Moore et al. 2006
    Randomised Control    trial
    2b
    Removal 8/52

      N = 7
    Retention, N= 113

      Broken, N = 9

      Intact, N = 104
    5
    OMAS:

      N/E
    No statistically    significant difference
  
  
    Schepers et al 2014
    Cross Sectional    Cohort Study
    4
    Removal 8-12/52

      N = 44
    Retention, N= 12
    51
    OMAS 73.22 vs 72.92    (p = 0.971)
    No statistically    significant difference
  
  
    Tucker et al. 2013
    Cross Sectional    Cohort Study
    4
    Removal 10-12/52

      N = 43
    Retention, N= 20
    31
    OMAS 75 vs 81.5 (p =    0.107)
    No statistically    significant difference
  
  
    Weening et al. 2005
    Cross Sectional    Cohort Study
    4
    Removal >9/52

      N = 30
    Retention, N= 20
    18.1
    OMAS: N/E
    No statistically    significant difference



Table 1:

There have been numerous biomechanical studies looking at various techniques and methods of fixation. In terms of material used, there does not appear to be a significant difference between titanium, stainless steel of bioabsorbable screws [3]. There does not appear to be a significant difference, biomechanically in terms of level of placement [4], three cortical vs 4 cortical fixation [5] or indeed 3.5mm vs 4.5mm screws [6,7]. There is, however, a biomechanical advantage to two vs one screw fixation [8] particularly in more proximal fracture patterns, i.e. Weber C variants. The biggest prognostic indicator of the efficacy of fixation remains proper anatomical reduction of the syndesomsis prior to fixation, with up to 52% of cases having inadequate reduction on post-operative CT [9,10].

Irrespective of the above, a common technical consideration for syndesmotic screw fixation is whether to remove or retain screws post fixation. Historically, screws were considered to hamper full movement across the tibiofibular articulation as well as potentially restricting full dorsiflexion. This was proposed due to an impairment of normal fibular rotation occurring secondary to syndesmotic widening. Later studies have suggested that this may not be the case, irrespective of retention of either intact or broken screws. This systematic review aims to look in more detail at the need for removal of screws and consider what the best practise is going forward.

Materials and Methods

A systematic review was conducted on the 10/03/2020 using PRISMA guidelines [11].

The following search terms were used: Ankle fractures; syndesmosis; ankle ligament; bone screws, ankle fixation; transyndesmosis; removal; retention.

These were used in the following databases: Embase, Medline and Conchrane library.

Exclusion criteria: not written in English, duplicate articles.

Inclusion criteria: articles containing ankle syndesmosis fixation with screws with direct comparision of retention vs removal.

These were then scrutinised to evaluate to exclude any articles which did not include Patient Related Outcome Measures (PROM) or review articles and case reports.

Results

The search yielded a total of 512 papers. Of these, 140 were duplicates across the search engines and were removed. Of the remaining articles the above inclusion and exclusion criteria were applied leaving 47 papers in total. These papers were then fully scrutinised as above leaving a total of 17 papers for use in the systematic review.

The papers were then reviewed to ascertain their level of evidence, according to the Oxford Level of Evidence criteria (OLOE). Each was then investigated to reveal the total numbers of both the retention and removal arms. Mean follow-up was also analysed as was mean outcome, looking at functional outcomes via either standardised OMAS (Olerud-Molander Ankle Score) or AOFAS (American Orthopedic Foot Ankle Score) scoring, bar one study where Baird and Jackson scores were available. The Conclusion, looking for functional differences across arms was also scrutinised.

Comparison of OMAS scores: This was discussed in 9 studies but the actual numbers were only provided for 6. The overall primacy differed between studies, with 5 demonstrating favourable results for screw removal a, 2 favouring retention and the others not providing numbers, but stating that the outcomes were not significant. The only paper that demonstrated a statistically significant result (miller) favoured removal, although the exact figures are not revealed.

Figure 1: PRISMA flowchart of studies identified, screened and included.

    
    
    Figure 1: PRISMA flowchart of studies identified, screened and included.

Comparison of AOFAS scores: These were provided in 8 studies. Across the board, there were no statistically significant differences between the removal and retention arms. In terms of raw numbers, 3 favoured removal, 4 favoured retention and 1 had no difference across arms. One study (Francesco) had results that functional results that contradicted each other, with OMAS favouring removal and AOFAS vice versa.

Comparison of Baird and Jackson Scores: Only one paper provided this score, favouring removal but with no statistical significance.

Discussion

The decision on whether to remove or retain syndesmosis screws remains a contentious topic. Proponents of screw removal argue that the syndesmosis should have healed by 12 weeks [12] and therefore removal of the screw will allow anatomical restoration of the fibular rotation and syndesmosis widening during the gait cycle.

The reverse arm argues that, given the above is the case, the screw will most likely break at the point where it restricts these movements and not in cases where there is no ongoing restriction and therefore removal is deemed unnecessary [15].

Given that removal involved further operative intervention, with complication rates of up to 22.4% [13], an argument can be made that this should only be undertaken with evidence to substantiate it. Even without the inherent risks of further intervention, the second operation also adds an economic burden [14].

With this in mind the above data suggests that in the majority of cases there is no indication for the routine removal of syndesmosis screws, from a purely functional point of view. Whilst there are radiological differences, such as diminished tibiofibular clear space, with narrowing in some cases of retained screws [15], this does not correlate to a functional impairment.

With this in mind, this review does not support the routine removal of syndesmosis screws postoperatively. The caveat to this remains patients who remain symptomatic through screw breakage or loosening, but these must be judged on a clinical basis and do not fall under the remits of routine removal.

Conclusion

Whilst controversy exists, this exhaustive and up-to-date analysis of the literature demonstrates there is no significant difference across both arms. When this is compounded with the added burdens, both financial and in terms of complications, there is evidence to suggest that routine removal of syndesmosis screws should not be undertaken and that this should be reserved for symptomatic patients only.

References

Download PDF

Citation: Shaunak S, Ahmed N and Singh S. A Systematic Review of the Routine Removal of Syndesmosis Screws in Traumatic Ankle Fractures. Austin J Orthopade & Rheumatol. 2022; 9(2): 1113.

Home

Journal Scope

Editorial Board

Instruction for Authors

Submit Your Article