Staged Management of Missed Lisfranc Injuries: A Report of Short‐term Results

Abstract

Objective

Lisfranc joint injury is a rare injury and can be easily missed at the initial treatment. Once ignored, late reduction is very difficult and requires extensive dissection. Surgical outcome is not as good as in the case of an early reduction. The aim of this cohort study was to analyze the midterm clinical and radiographic outcomes of staged reduction and fixation in a consecutive series of patients with old Lisfranc injuries.

Methods

Fifteen patients (16 feet) with missed Lisfranc injuries were treated with staged reduction. Mean duration between injury and surgery was 4.8 months (3–8 months). In the first stage, an external fixator was applied across the Lisfranc joint or/and Chopart joint and distraction was done at 1–2 mm/day. In the second staged, open reduction and internal fixation (ORIF) were done and we were able to reduce all the fractures and dislocations.

Results

The mean duration between two surgeries was 3.2 weeks (range, 2.5–4.5 weeks). Anatomic reduction was obtained in all 15 patients. At the last follow‐up point, 7 feet had good functional results, 5 feet fair, and 4 feet poor functional results. In the 4 patients who achieved poor functional results, 2 cases were due to severe injuries to the articular surface and tissue scaring; 2 cases were due to loss of reduction. For the 4 feet with poor functional results, 2 were scheduled for secondary arthrodesis during the follow‐up. The average American Orthopaedic Foot and Ankle Society Midfoot Scale (AOFAS) scores for these patients were 75.8 points (range, 43–98 points). The pain visual analog scale (VAS) was 3.1 points at the final follow‐up.

Conclusion

Our study demonstrated that staged reduction and extra‐articular fixation should be considered for old Lisfranc injuries with a good reduction, firm stability, and low risk of intraoperative fracture and soft tissue complications.

Introduction

Lisfranc joint injuries (tarsometatarsal [TMT] dislocations or fracture dislocations) are relatively rare, making up a reported 0.1%–0.9% of all fractures, and occur in 1 person per 55,000 yearly1, 2. However, 20%–40% of patients are not treated in a timely manner as these injures are either overlooked or misdiagnosed as foot sprains or isolated fractures of the tarsal and metatarsal bones by the initial treating physicians3, 4. The injury can be overlooked in patients with multiple traumas and in cases where the diastasis spontaneously reduces. This percentage is still higher in developing countries due to lack of awareness about this injury, inadequate diagnostic evaluation, or patients not seeking medical care. If overlooked or not treated correctly, Lisfranc joint injuries frequently result in chronic pain and functional loss due to residual ligamentous instability, deformity, or/and arthritis; osteoporosis also may occur because of long‐term antalgic gait without weight‐bearing.

The surgical principles of treating symptomatic missed Lisfranc injury include anatomic alignment of the midfoot joints and restoration of normal ligamentous stability by stable internal fixation5, 6. Misdiagnosis of Lisfranc injuries, unstable fixation, inadequate reduction, and failed conservation treatment may result in residual deformities and, ultimately, pain. For obtaining a minimally painful functional foot, some treatment options for missed Lisfranc injury have been proposed, such as open reduction and internal fixation, TMT joint realignment with joint preservation, arthrodesis, and arthroplasty. The treatment method and treatment effect may depend on the degree of deformity, the number of affected joints, soft tissue injury, articular cartilage loss, chondrocyte damage, and ligamentous injury. As for acute Lisfranc injuries, the best results in the treatment of missed Lisfranc injuries are achieved with stable anatomic reduction. Theoretically, anatomic reduction and rigid internal fixation should be applied in these injuries to decrease the risk of recurrent dislocation, subsequent arthritis, and further damage of articular cartilage in old Lisfranc injuries7, 8. Arthrodesis also may be a better choice for the patients with purely ligamentous injuries or massive articular damage than joint realignment and ligament reconstruction for reducing the risk of recurrent subluxation and progressive arthritis.

The literature has little information about the timing of open reduction internal fixation of Lisfranc injuries when the injury is no longer acute. Much of the controversy associated with these injuries is based on the timing of surgery. For missed Lisfranc injuries, some authors report that good results have been achieved with open reduction internal fixation performed up to 6 weeks after injury9, 10. If open reduction and internal fixation are performed within 6 weeks of Lisfranc injury, outcome scores of these patients are similar to those who have undertaken acute surgery. Over a period of 6 weeks, the results of open reduction and internal fixation for Lisfranc injury may be poor secondary to extensive soft tissue dissection, poor stabilization because of rounding of the ligament edges, and articular destruction caused by joint malalignment11, 12.

For decreasing potentially significant sequelae occurring in missed Lisfranc injuries, it is important to gain anatomic alignment of the midfoot joints and restoration of normal ligamentous stability. Significantly worse outcomes have been shown when operative treatment has been delayed. Late reduction of TMT dislocations is difficult because of scarring and contraction of soft tissues, osteoporosis, articular cartilage damage, and remodeling of tarsal bones8, 13, 14, 15. Although anatomical reduction is the most important predictor of good or excellent results after tarsometatarsal fracture dislocations, treatment of these missed injuries with open reduction and internal fixation would theoretically provide better functional results as compared to salvage procedures. However, this often requires extensive soft tissue releases to achieve adequate reduction, leading to extensive loss of vascularity of tarsal bone and soft tissue complications; in addition, when correcting deformity, intraoperative fractures often occur because of osteoporosis. These complications seriously affect the prognosis of the injured foot.

For achieving the best anatomic realignment and stabilization of the Lisfranc joints, avoiding excessive soft tissue dissection, our study patients underwent a staged treatment in which soft tissues and Lisfranc joint are gradually distracted with a spanning external fixator followed by open reduction and internal fixation. We propose this alternative surgical procedure for treating missed Lisfranc injuries and hypothesize that staged management of these injuries is better than single‐stage surgical release. In this study, we assess the short‐term (mean, 2.2 years) clinical and radiographic outcomes of missed Lisfranc injuries treated with this strategy.

Materials and Methods

We performed a retrospective study of patients with missed Lisfranc injuries who were treated in the West China Hospital between May 2011 and May 2014 at a mean follow‐up of 18 months (1.5–33 months) after the initial injury. This retrospective study was approved by our institutional review board. All the treatment procedures for the included patients were carried out under the guidance of a single surgeon, the senior author (H. Z.); the patients were followed up for at least 1 year.

Inclusion and Exclusion Criteria

The inclusion criteria were: (i) patients presented with residual deformity or pain after Lisfranc injuries and accepted staging treatment strategy; and (ii) the time interval from initial treatment to the final surgery was more than 6 weeks. The exclusion criteria: (i) Patients who had neurological disorders or pre‐existing deformity of the foot; (ii) Patients who did not give their consent.

Study Population

Twenty‐three patients (24 feet) were identified from the trauma registry database. During the follow‐up period, 1 patient died as a result of myocardial infarction, 3 were lost to follow‐up, 3 patients were diagnosed with severe post‐traumatic arthritis and had undergone arthrodesis, and 1 patient refused to participate in the study. Finally, 15 patients (16 feet), including 6 women (6 feet) and 9 men (10 feet) with an average age of 38.7 years (range, 25–46 years), were available for the clinical and radiographic evaluation. The mean height and weight of the patients was 165 cm (range, 158–180 cm) and 57 kg (range, 45–74 kg), respectively. The time interval from the initial treatment to the final surgery ranged from 1.5 to 33 months (mean, 18 months). All patients gave informed consent for participation, and the Human Subjects Research Committee approved the study.

Most of the injuries were caused by high‐energy trauma. Five patients were injured in a traffic accident, 7 were injured by striking, and 3 were injured as a result of falling from a height. Four patients had multiple injuries at the time of the accident, 3 of which had an injury to the mid‐tarsal (Chopart) joint. Five patients had opened Lisfranc injuries; 10 patients had closed Lisfranc injuries. The diagnosis of 5 patients had been missed on first presentation. The remaining 10 patients had received primary treatment: 5 patients were treated with open reduction and K‐wire fixation, 3 patients received treatment with closed reduction and K‐wire fixation, and 2 patients received treatment with reduction with immobilization in a cast walker. The reason for seeking treatment again was pain in 7 cases and deformities with pain in 8 cases.

By assessing preoperative radiographs and confirming intra‐operatively. The fracture classification was categorized as follows: 8 patients had homolateral injuries; 4 patients had isolated injuries; and 3 patients had divergent injuries. Lisfranc joint injuries were classified according to a system described by Myerson et al.13.

There were 15 patients (16 feet) included in the study: 9 male and 6 female. The mean (range) injury age was 38.7 years (25–46 years). The causes of the injury were traffic accidents (5 cases, 5 feet), falls from height (3 cases, 4 feet), and striking injuries (7 cases, 7 feet). The fracture classification could be divided into three types: total incongruity (8 cases), partial incongruity (4 cases), and divergent (3 cases). Metatarsal fracture (7 cases), cuboid fracture (1 case), navicular fracture (2 cases), and cuneiform fracture/dislocation (6 cases) were the associated foot lesions. There were 3 cases of ipsilateral lower limb multiple fracture, 1 case of contralateral lower limb multiple fracture, and 4 cases of polytrauma/multiple injuries. The duration of follow‐up was 2.2 years (1–3 years).

Surgical Technique

All patients were treated in two stages. In the first stage, a spanning external fixator was applied across the Lisfranc joint with one full pin through the calcaneus and tales, and two half pins fixed the calcaneus from the side. At the same time, one full pin across the 1–5 metatarsal bone and one half pin through the base of the first metatarsal bone. If combined Chopart joint dislocation, talus neck, navicular, and cuboid were fixated with Kirschner wires. The Lisfranc and chopart joints were tracted simultaneously. One‐stage traction operations were performed in the patients with osseous connection after osseous structure osteotomy. When the length of the medial and lateral column returned to normal even excess traction, second‐stage operations were performed to achieve articular reduction and fixation.

The external fixator was used for gradual distraction across the tarsometatarsal bones at the rate of 1 mm/day. Patients were followed up with weekly radiographs after surgery to judge adequate distraction (Fig. 1). During the immediate postoperative period, traction was applied, which speed of which can be increased up to 2–3 mm/day. The patients should be encouraged to engage in partial weight‐bearing walking as early as possible, avoiding articular reduction; strong fixation in the second‐stage was affected by osteoporosis. At the same time, the patients' perioperative analgesia treatment should be evaluated.

Figure 1.

First stage of reconstruction with a spanning external fixator. Tarsometatarsal bones are gradually distracted to restore length and alignment using the fixator. (A) Lateral appearance; (B) anteroposterior appearance; (C) lateral radiograph; and (D) anteroposterior radiograph.

When the length of the medial and lateral column recovery even harper‐traction by radiographic assessing, a second‐stage open reduction and internal fixation using transarticular screws and plates would be scheduled. A longitudinal incision is made over the dorsomedial aspect of the first TMT joint. The incision should be long enough to allow for visualization of the second TMT joint and the Lisfranc area. Care is taken to identify and protect the medial branch of the superficial peroneal nerve. Dissection is then carried out in the interval between the extensor hallucis longus and the anterior tibialis tendons. The periosteum is incised longitudinally. The periosteum is elevated medially and laterally to allow adequate visualization of the TMT joint. The extensor hallucis longus is retracted laterally, and dissection is carried out plantar to the tendon. Care must be taken not to damage the dorsalis pedis artery and deep peroneal nerve to avoid injury. Once all involved medial joints are visible, a thorough debridement of the joints is performed to remove any interposed tissue. Reduction and fixation of the injury is carried out in a medial‐to‐lateral manner, starting with the first TMT joint. Once adequate reduction is confirmed, the reduction can be held manually. A precontoured first TMT plate may be used. The plate is brought into the field and positioned over the first TMT joint. Care is taken to avoid injury to the anterior tibialis tendon. When the proper plate position is achieved and verified with fluoroscopy, the plate is held in place with a plate tack proximally. Drilling and screw fixation of the metatarsal is then performed. With the first TMT stabilized, the intercuneiform articulation is then addressed. If instability is present at this joint, reduction is performed in a standard manner and held in place with a pointed reduction clamp. The most proximal of the screw holes on the plate is then drilled in an intercuneiform manner and replaced with a fully threaded cortical screw. The medial cuneiform second metatarsal articulation is then addressed. Reduction is carried out in a manner similar to the above and held in place with a pointed reduction clamp. Fluoroscopic guidance is then used to place the Lisfranc screw. A partial threaded 4.0‐mm cannulated screw or fully threaded cortical screw is placed through the plate. In the event that the lateral rays are also involved, indirect or direct reduction of these joints is performed and fixed with transarticular 0.062 K‐wires. Alternatively, a first interspace incision may be used with dorsal plating of the first TMT joint. In this case, a locked plate is used and the Lisfranc screw is placed percutaneously outside the plate (Fig. 2a–f). Determining which incision to use is based on whether a full evaluation of the second TMT joint is needed or not, as visibility is better with this approach.

Figure 2.

Preoperative radiograph showing a missed Lisfranc injury. (A) Anteroposterior, (B) lateral views and (C) oblique views. Postoperative radiographs showing three extra‐articular plates that were used and a cannulated transarticular screw that was placed percutaneously outside the plate to fix the medial and middle rays. The lateral rays were fixed by K‐wires. (D–F) Anteroposterior, lateral, and 45° oblique radiographs on weight‐bearing views.

Postoperative Management

When returned to the resuscitation room, immediate measures were taken for all patients to prevent or reduce postoperative swelling of the limb, including elevating the limb, immobilization, and cryotherapy. Wounds were examined 1 week later. A short leg splint was worn for at least 2 weeks for immobilization, and a removable bivalve cast was used for 4–6 weeks without weight bearing. At 2 weeks postsurgery, partial weight‐bearing of 10 kg was allowed with a removable bivalve cast. Between 8 and 12 weeks after surgery, partial weight bearing would increase gradually; a removable bivalve cast and physical exercises (gait training and range‐of‐motion exercises) would start. At the third month, it was suggested that the patients ambulate in a stiff‐soled shoe, moving gradually to the point of full weight bearing.

Follow‐up was performed 3 days postoperatively, every 1 month during the first half year, every 3 months during the first 2 years, then annually. All patients had been followed up for at least 1 year.

At the end of the follow‐up, we assessed clinical outcomes using the American Orthopaedic Foot and Ankle Society (AOFAS) Midfoot Scale16. The clinical outcomes were divided into four grades as follows: excellent (>90 points), good (80–89 points), fair (70–79 points), and poor (<70 points). The condition and healing time of incision were also recorded.

Alignment was assessed using lateral anteroposterior weight‐bearing radiographs and 45° oblique radiographs. The standard of anatomic reduction was that the lateral edge of the base of the first metatarsal bone and the medial edge of the base of the second metatarsal bone were within 2 mm, the medial border of the second metatarsal lined up with the medial border of the middle cuneiform on the anteroposterior radiograph, and the medial border of the fourth metatarsal lined up with the medial border of the cuboid on the oblique radiograph17. The arch height was defined as the distance between the base of the fifth metatarsal and cuboid in the lateral radiograph. The talus‐first metatarsal angle formed by the longitudinal axis line of the talus and the first metatarsal bone was also used to evaluate sagittal deformity14. These parameters were compared with the normal foot. Posttraumatic osteoarthritis was diagnosed if any evidence of osteophytes, joint space narrowing, or subchondral cysts or sclerosis in conjunction displayed on weight‐bearing radiographs along with the related clinical symptoms.

Statistical Analysis

The normality of the continuous data was checked with the Kolmogorov–Smirnov test. Normally distributed variables were compared using paired Student's t‐test and non‐normally distributed variables were compared using the Wilcoxon signed‐rank test. Pearson's χ²‐test was used for analysis of categorical data; if there were more than 20% cells with expected counts below 5 or the expected count was below 1 in any cell, Fisher's exact test was used instead of Pearson's χ²‐test. A P‐value of 0.05 or less was considered statistically significant.

Results

The average duration of follow‐up was 2.2 years (range, 1.0–3.0 years).

Complications

All patients were observed as having too much callus and cicatricial tissue forming in the injured foot, which required a lot of tissue excision. One patient complained of abnormal sensation of the dorsum of the foot after surgery, which indicated a lesion of the superficial peroneal nerve. One patient presented superficial wound edge necrosis on the third post‐operative day, but the wound eventually healed. Two patients felt moderate pain after a long period of walking, with mild lateral subluxation observed in one and medial subluxation in the other on weight‐bearing radiographs; posttraumatic arthritis developed in both of these patients. One was satisfied with the result and declined further surgery but the other patient underwent arthrodesis because of persistent pain at the end point of follow‐up. One patient developed abduction deformity when weigh‐bearing walking, because the medial column was not fixed. No implant failure and delayed union or nonunion of fractures were noted, and no painful neuromas, deep‐vein thrombosis, or iatrogenic vascular injury occurred.

Radiographic Results

All the patients achieved anatomical reduction according to measurements of radiographic parameters on the first post‐operative day, and during the follow‐up period, there were no patients observed with loss of internal fixation. However, loss of reduction was observed in 2 patients; they had pure ligamentous injuries. We observed posttraumatic osteoarthritis in 5 patients (5 feet, 31.3%), which included the 2 patients with loss of reduction.

The parameters of the talus‐first metatarsal angle in anteroposterior and lateral in weight‐bearing radiographs, the distance between the base of the first and second metatarsals in anteroposterior weight‐bearing radiographs, and the distance between the base of the fifth metatarsal and the cuboid in lateral weight‐bearing radiographs were measured at the final follow‐up, and the values were close to normal. When compared with the normal foot, no statistically significant differences were found. Detailed demographic data are shown in Table 1.

Table 1.

Summary of main radiographic results (mean ± SD [range])

Sides	TFMA (anteroposterior view, °)	TFMA (lateral view, °)	MT1‐2 distance (mm)	Cuboid – MT5 distance (mm)
Normal feet	5.2 ± 2.4 (−3–11)	0.2 ± 1.5 (0–2)	1.3 ± 2.2 (0–4)	5.8 ± 3.2 (0–14)
Injured feet	5.8 ± 3.7 (−5–16)	0.3 ± 3.6 (−4–12)	1.4 ± 2.3 (0–4)	5.7 ± 3.5 (0–13)
P‐value	0.590	0.346*	1.000*	0.933

MT1‐2 distance, distance between first and second metatarsal base; NS, not significant; TFMA, talus first metatarsal angle.

^*Wilcoxon signed‐rank test due to skewed distribution.

Clinical Results

At the end point of follow‐up, the average AOFAS scores for these patients was 75.8 points (range, 43–98 points), which was mainly influenced by mild pain and decreased recreational function. Compared with the average AOFAS score of 34.6 (0–58) before surgery, there was a substantial improvement in the scores after surgery. According to the AOFAS scores, 7 feet had good functional results, 5 feet had fair functional results, and 4 feet had poor results. In the 4 patients who achieved poor functional results, 2 were due to severe injuries to the articular surface and tissue scaring, and 2 were due to loss of reduction. In the 4 poor functional feet, 2 were scheduled to secondary arthrodesis during the follow‐up.

After conducting a subgroups analysis, we found no difference in AOFAS midfoot scores among different types of foot injuries, such as cuneiform fractures, cuboid fractures, multiple injuries, and ipsilateral lower limb multiple fractures. However, those patients without loss of reduction had a higher average AOFAS score than those patients with loss of reduction, and those with a shorter delay time for surgery also had a higher average AOFAS score (Table 2).

Table 2.

Comparison of the American Orthopaedic Foot and Ankle Society Midfoot Scale (AOFAS) scores within subgroups

Subgroups	Number of patients	AOFAS score
		Mean ± SD (points)	P‐value
Pure ligamentous injury	3	58.8 ± 10.6	0.136
Combined ligamentous and osseous injury	12	76.6 ± 18.3
Cuneiform and/or cuboid injury	8	72.0 ± 18.6	0.808
No cuneiform or cuboid injury	7	74.2 ± 15.2
Isolated injury	4	79.8 ± 22.9	0.414
Multiple injuries	11	70.5 ± 17.5
Multiple trauma	4	66.8 ± 13.7	0.374
No multiple trauma	11	75.3 ± 16.4
No loss of reduction	13	75.9 ± 13.1	0.047
Loss of reduction	2	54.5 ± 7.2

Discussion

Approximately 20%–40% of Lisfranc injuries can be misdiagnosed or simply missed because of the lack of awareness about this injury, inadequate diagnostic evaluation, or patients not seeking medical care12. If Lisfranc injury is missed or the primary treatment fails as a result of faulty anatomical reduction, implant failure, or invalid conservative measures9, tarsometatarsal fractures and dislocations frequently result in painful malunion and impaired function with a broad spectrum of deformity11, 18, 19, 20, 21. Over a period of 6 weeks, the results of open reduction and internal fixation for Lisfranc injury may be poor secondary to extensive soft tissue dissection, poor stabilization because of rounding of the ligament edges, and articular destruction caused by joint malalignment11, 12. In this study, when the treatment is delayed for over 6 weeks, we observed that patients presented with painful arthritis or residual deformity after tarsometatarsal fracture dislocation, so we defined these Lisfranc injuries as missed/old injuries.

Initial non‐operative treatment was recommended when old Lisfranc injuries met the standard as follows: there were not associated unstable fractures and significant TMT subluxation or instability, or there was significant TMT subluxation or instability with TMT joint arthritis7. Some other authors proposed that anatomical reduction was the most important predictor of good or excellent results after TMT fracture dislocations22, 23, 24; initial conservative treatment may obtain inadequate reduction or unstable fixation, and this can result in gross deformity, painful disability, and undue socioeconomic costs1, 25, 26. In our study, 5 patients had been missed on first presentation, 5 patients were treated with open reduction and K wire fixation, 3 patients were treated with closed reduction and K wire fixation, and 2 patients were treated with reduction with immobilization in a cast walker. We agree that closed reduction is not sufficient to achieve anatomical reduction in the majority of Lisfranc injuries because of interposed soft tissue and small bony fragments from fractures of the base of the metatarsals. K‐wire fixation or immobilization in plaster alone cannot provide rigid fixation and maintain reduction, which may induce a higher failure rate in patients with unstable joints9, 10.

Late reduction of old Lisfranc injuries is very difficult because of the presence of old malunion fractures, stiffness and contraction of the soft issues and Lisfranc ligament, osteoporosis, and (or) arthritis8, 13, 14, 15. Therefore, we propose that staged reconstruction is more suitable for managing old Lisfranc injuries than single‐stage surgical release. Soft tissues and tarsal bones are gradually distracted with a spanning external fixator. This strategy requires a longer treatment cycle and costs more money, but the extensive dissection and soft tissue stripping at the time of surgery are avoided; nerves and blood vessels can also be protected. Besides, the distractor protects the tarsal bones from excessive and non‐anatomic forces. Thus, we believe that staged reduction is an effective and safe method for old Lisfranc dislocations.

Arthrodesis is an important treatment option for old Lisfranc injuries with advanced arthritis or fixed deformity or neuropathic deformity7. In terms of reducing the risk of recurrent subluxation and progressive arthritis, arthrodesis may also be a better choice than joint realignment and ligament reconstruction for patients with purely ligamentous injuries or with massive articular damage27, 28. For patients with old dislocations, re‐dislocation or deformation because of the absorption of ligament or contracture, in our opinion, medial column arthrodesis in one stage may be a better choice. Komenda et al. reviewed 10 patients who underwent TMT arthrodesis for intractable pain after a traumatic midfoot injury; the diagnosis was missed initially and the average time from injury to surgery was 34 months (range, 6–81 months)29. After arthrodesis, their average AOFAS midfoot score increased from 41.1 to 82.2. Although arthrodesis may relief pain and lead to a better AOFAS score, it is at the cost of joint motion and can make the normal associated joint stiff, especially in younger, more active individuals. In our practice, not all Lisfranc injuries are treated with arthrodesis. For Lisfranc injuries treated within 3 months, if they have severe arthritis, arthrodesis may be performed. Even if Lisfranc injuries are more than 3 months old upon diagnosis, TMT has subluxation or instability with mild–moderate arthritis and intact articular surface, open reduction and rigid internal fixation is recommended to keep the function of joints and decrease the risk of severe arthritis. In our study, 2 patients with pure ligamentous injuries presented loss of reduction followed by arthritis during follow‐up. These patients account for a significant proportion of all patients with arthritis (40%). This result supports the concept of Kuo et al.10, they point out that patients with purely ligamentous injuries have a tendency to a higher prevalence of posttraumatic osteoarthritis compared with patients with combined ligamentous and osseous injuries osteoarthritis. In these patients with pure ligamentous injuries, primary arthrodesis may be a better choice. Considering that primary arthrodesis requires more extensive dissection, including removal of viable bone to prepare the articular surfaces which increases the potential for joint instability and necessitates larger bone grafts6, for those joints with intact articular surface on preoperative radiographs and confirmed intraoperation, we do not recommend primary arthrodesis in these patients.

Internal fixation is used for achieving stabling of bony structure, but the most appropriate approach for internal fixation is still controversial. At initial treatment, the major method for fixation is Kirschner‐wire fixation9, 10. However, many complications have been reported with the use of Kirschner wire, including pin migration, pin tract infections, and loss of reduction18. Biomechanical evidence shows that screw fixation provides a stronger and more stable construct than Kirschner wire30. However, screws also have defects, such that a second operation can be required to remove the screws; the use of bio‐absorbable screws may be beneficial, but the risk of breakage, loss of fixation, and articular damage cannot be avoided when these screws degrade31. Moreover, screws can cause secondary damage to the articular cartilage by transarticular fixation32. Therefore, some researchers suggest that plates present an acceptable alternative to be used for internal fixation of Lisfranc injuries33. Alberta et al. conducted a biomechanical comparison of dorsal plate and trans‐articular screw fixation in 10 matched pairs of fresh‐frozen cadaver lower extremities, and found no significant difference between plates and screws in resisting tarso–metatarsal joint displacement with weight‐bearing load, but the area of articular surface destruction caused by screws ranged from 2.0% to 4.8%32. This “second hit” may potentially accelerate the posttraumatic arthritic process, and the amount of arthritis is related to the area of damage on the articular surface. Primary fixation using dorsal plates to avoid further articular damage is recommended. Although severe wound complications did not occur in our study, using plates might increase the risk of such complications.

Conclusions

Our study has demonstrated that staged reduction and extra‐articular fixation is an effective method for treating old Lisfranc injuries. Staged reduction and extra‐articular fixation should be considered for old Lisfranc injuries with a good reduction, firm stability, and low risk of intraoperative fracture and soft tissue complications.