1 Introduction

Midfoot injuries are relatively uncommon with an estimated incidence of approximately 3–12/100,000 per year,^1,2 with injuries to the tarsometatarsal joints being the most common encompassing a range of pathologies.³ Classification of tarsometatarsal joint injuries has been attempted by a number of authors, with Quenu and Kuss, describing three patterns of injury in 1909,⁴ which formed the basis of the classification by Hardcastle and Myerson.^5,6 Schepers and Rameldt classification describes the injuries by columns, the medial column is comprised of the first metatarsal and medial cuneiform, the middle column includes the second and third metatarsals and the intermediate and lateral cuneiforms, and the lateral column consists of the fourth and fifth metatarsals and the cuboid (Fig. 1).⁷ Despite central column and Lisfranc injuries being the commonest midfoot injury a large proportion, up to 20 %, are initially missed and therefore this number may be under reported,⁸ there is little known on the proportion of lateral column injuries that are missed.

Close

Fig. 1 Illustration showing the anatomical three columns described by Schepers and Rameldt.

Export to PPT

The tarsal navicular joint is the keystone of the medial column and bears the majority of the load in the tarsal complex during weightbearing.⁹ Previous literature has centred on the medial column and specifically Lisfranc injuries. Although there have been reviews of the management of cuboid fractures specifically^10,11 as part of the lateral column there is little reported about the importance and the impact of lateral column injuries as a whole. The literature regarding the lateral column at the tarsometatarsal articulation is absent at present.

The principles of tarsometatarsal injury management are ubiquitous in that the primary aim is anatomical reduction, and midfoot stabilisation. Lateral column injuries have historically been treated with Kirschner wire fixation when encountered, however management of these injuries does vary from non-operative, percutaneous and open fixation. When open fixation is performed there are various methods of fixation used to achieve reduction. To date there has been no direct comparison of modes of lateral column injury fixation and outcomes.

2 Aim

Our aim in this study was to analyse lateral column injuries to the midfoot, their method of treatment and their radiological outcomes. Our null hypothesis being that fixation is required to obtain and maintain lateral column alignment.

3 Methods

This was a retrospective observational study of midfoot fracture/dislocations in four foot and ankle units in the UK, two level 1 major trauma centres and two level two trauma units. The collaboration between units was to enable collection of sufficient volume of cases. The protocol was reviewed by the Liverpool Orthopaedic and Trauma Service research review board and was evaluated to be a service evaluation project and therefore did not require ethical approval. All surgically treated midfoot injuries were collected and analysed for potential inclusion into the study. Data was collected between January 2011 and August 2022. Inclusion criteria for this study was tarsometatarsal joint injury that included the lateral column, which were followed up for a minimum of 6 months with repeat radiographs. Exclusion criteria included patients under the age of 16 and mangled foot injuries.

Patients medical records and radiological imaging were reviewed and demographic data collected. Radiological images were reviewed using departmental imaging software (Vue PACS, Carestream, Version 11.4.1.0324). Anonymised data was collected locally in each unit on prepared spreadsheets and sent to the lead centre for further analysis. The primary outcome of the study was lateral column reduction and maintenance of reduction to six months post-surgery. Other possible contributing factors to the primary outcome were recorded, included demographics, polytrauma, columns injured as described by Schepers and Rammelt, column fixation methods and mechanism of injury.¹²

4 Statistics

Continuous parametric data are presented as the mean and 95 % confidence intervals, dichotomous data as cross tabulation of frequencies and percentages. Statistical analysis was performed using student t-test if continuous data was tested to be normal and Mann Whitney or Fishers Exact test if tested to be non-normally distributed. Binary data was tested using Chi square. Uni- and multivariate analyses were performed using univariate and multivariate logistic regression analysis to identify factors involved in lateral column stability. The univariant analysis was performed using lateral column alignment as dichotomous dependent variable. Any factor which achieved significance on univariate analysis underwent further multivariant regression analysis. Significance was given to variables that reached p < 0.05. Statistical analysis was undertaken using SPSS statistics version 26 (IBM, New York, USA).

5 Results

A total of 409 surgically treated midfoot injuries were identified for further investigation. Following analysis, a total of 235 cases were diagnosed as having a lateral column injury and could be included in the study. Of the 235 cases, 143 (60.9 %) were male and 92 (39.1 %) were female. There were 183 (77.9 %) cases which were single limb injuries and 52 (22.1 %) associated with polytrauma. All but 1 case (234, 99.6 %) of lateral column injury was associated with central column injury. Lateral column injury was also associated with medial column injuries in 166 cases (70.6 %). Purely ligamentous injury was a minority, occurring in 16 cases (6.8 %). All 4 units had variable fixation methods with 1 unit preferring to fix the lateral column on all cases, and some units fixing the lateral column in only a few (Table 1).

Of the 235 lateral column injuries, data was available regarding fixation and radiographic alignment at surgery and at least 6 months post-surgery in 222 cases (Table 2). There were 44 cases where the lateral column underwent Kirschner wire fixation, 23 lateral column plate fixations and 3 lateral column screw fixations. Alignment of the lateral column at time of surgery and at final radiograph are shown in Table 1. The average time to final radiograph was 77.52 weeks (95%CI 64.97, 90.07).

Both Univariate and Multi-variate regression analysis was performed to identify possible factors in the cases where there was loss of lateral column alignment at the final radiographs. This is shown in Table 3. The use of a bridge plate to fix the central column appears protective and purely ligamentous injury was a higher risk than an injury that included the bone. There were only 7 cases (3.18 %) where the lateral column alignment was lost at follow up. Given the increase in loss of lateral column alignment in the K wire fixation group, in an experimental model this would result in a negative numbers needed to treat (NNT) scenario. For plates and screws, the NNT to prevent a loss of alignment event would be 35.

6 Discussion

Our aim in this study was to analyse lateral column injuries to the midfoot, their method of treatment and their radiological outcomes. A total of 409 patients across the four units were identified as having any form of midfoot injury requiring fixation with 235 patients (57 %) from this cohort being diagnosed as having a lateral column injury, which is in keeping with prevalence of lateral column injuries in midfoot trauma previously reported.¹ In our study, 147 patients (63 %) had no fixation for their lateral column injury with only 2.84 % losing alignment at subsequent follow up. The patients undergoing K wire fixation had a greater loss of alignment rate (5.88 %), although this could be due to more continued instability cases in this group following central column stabilisation, as the alignment at time of surgery was taken from the final intraoperative imaging. This was found clinically, Fig. 2 shows the plain radiograph and CT images of a patient with multiple column injuries and midfoot instability. Upon initial temporary stabilisation of the medial column intraoperatively (Fig. 3A) there was still lateral column instability which required stabilisation (Fig. 3B) prior to open reduction and internal fixation with plates (Fig. 3C). Postoperative plain radiographs were comparable to intraoperative images in this case (Fig. 4). Nevertheless, it can be argued that K wires are not needed routinely to fix lateral column injuries and therefore our nul hypothesis can be rejected.

Close

Fig. 2 Initial injury AP, Lateral, and Oblique plain radiographs (A,B,C) and Sagittal CTcross-sectional imaging (D&E) showing multi column injury with midfoot instability.

Export to PPT

Close

Fig. 3 Intraoperative radiographs showing lateral column instability after initial medial column temporary fixation (A) necessitating the need for stabilisation of the lateral column with k wires (B) prior to medial column open reduction internal fixation (C).

Export to PPT

Close

Fig. 4 Postoperative AP, lateral, and oblique plain radiographs showing final medial column stabilisation with internal fixation and lateral column stabilisation with k wires.

Export to PPT

Univariate and Multivariate regression analysis on the loss of alignment of the lateral column at final follow up showed there were no statistical significant differences based on; patient demographics, concomitant medial or column central injury, mechanism of injury, or if an isolated injury or associated with poly trauma. The only significant factors were the presence of ligamentous injury and the use of central column bridge plating was protective. Theoretically these factors maybe due to an increase in central column malreduction in the fusion/transarticular compared to bridge plating group, and thus having a greater likelihood of malreduction and later loss of alignment in the lateral column.

In our study we have found that in the majority of cases (97.2 %), not fixing the lateral column will result in no loss of alignment. Therefore, we can be reassured that in the majority of cases where the lateral column alignment has been restored following central column stabilisation, no further fixation is required in the lateral column. Where fixation was undertaken, there was no loss of alignment if screws or plates were used. In the cases where K wires were used, at the time of surgery 3 out of 51 patients (5.88 %) did not have lateral column alignment after surgical fixation, this increased at the final 6-month radiological check to 6 patients (11.8 %). Therefore 3 patients (5.9 %) who had satisfactory lateral column alignment at the time of surgery subsequently lost this through K wire fixation. It is unknown if an increase in time in the use of wires (i.e. not removing them before greater than 6 weeks) would have improved the loss of alignment results.

Associated central column injury occurs in almost all cases (99.6 %). Previous cadaveric studies have shown a lateral suspensory ligament bridging from the central column to the lateral column, which may explain why stabilisation of the central column, most commonly also reduces the lateral column.¹³ Tensioning the suspensory ligament, restores the transverse metatarsal arch and thus normalises the lateral column alignment. There have been no other studies comparing central column fixation and lateral column stability to date. The sequalae of lateral column instability is not fully understood but has been associated with adult acquired flat foot deformity (AAFD),¹⁴ ligament instability¹⁵ and post traumatic arthritis.¹⁶

Our study has limitations. As a retrospective study we are unable to determine the reasons for stabilisation, although most surgeons involved felt that routine use was common even when normal lateral column alignment had bene established. There were much smaller numbers of plates and screws used, which means that we are unable to recommend their use instead of K wires, when alignment has not been established. A randomised control trial would allow the question to be tested if K wires are useful in patients where lateral column alignment has been established.

7 Conclusion

Lateral column injury occurs in over half of midfoot fractures in this study. It rarely occurs alone and is most commonly related to three column injuries. Nevertheless, following stabilisation of the central column, additional fixation of injuries to the lateral column are unlikely to be required in the majority of cases. In cases where lateral column stabilisation is required, plates and screws may be preferable to K wires. The use of a bridge plate to fix the central column appears protective and purely ligamentous injury was a higher risk of lateral column displacement than an injury that included the bone.

Funding

No external or funding or sponsorship was utilised in this project.

Ethical approval and patient consent

The protocol was reviewed by the Liverpool Orthopaedic and Trauma Service research review board and was evaluated to be a service evaluation project and therefore did not require ethical approval and information was pseudoanonymised so individual patient consent was not needed.

Financial Support and sponsorship

No external or funding or sponsorship was utilised in this project.

CRediT authorship contribution statement

Thomas R.W. Ward: Formal analysis, Writing – original draft, Visualization. Khalis Boksh: Investigation, Validation, Writing – original draft, Visualization. Grace Airey: Investigation, Validation, Visualization. Darren Myatt: Investigation, Validation, Visualization. Junaid Aamir: Investigation, Validation, Visualization. James Chapman: Investigation, Validation. Htin Kyaw: Investigation, Validation, Project administration. Lucky Jeyaseelan: Writing – review & editing, Project administration. Lauren Greasley: Investigation, Project administration. Isabella Drummond: Investigation, Project administration. Mamdouh Elbannan: Conceptualization, Methodology, Writing – review & editing. Hiro Tanaka: Conceptualization, Methodology, Formal analysis, Writing – review & editing. Jitendra Mangwani: Conceptualization, Methodology, Formal analysis, Writing – review & editing, Supervision. Lyndon Mason: Conceptualization, Methodology, Formal analysis, Writing – review & editing, Supervision.