Background

Flatfoot deformity involves multiple components of medial arch collapse, hindfoot valgus, and forefoot abduction [1, 2]. Pediatric flatfoot can be rigid or flexible. Flexible flatfoot may be symptomatic or non-symptomatic, with preserved medial arch with non-weight bearing and arch flattening during stance [3, 4].

Asymptomatic patients follow a natural course of improvement over time and require periodic monitoring for signs of progression. In contrast, symptomatic forms can present with foot, leg, or knee pain, tight Achilles tendon, and walking difficulties [5,6,7].

Initially, activity modifications and orthotics are prescribed. In addition, nonsteroidal anti-inflammatory drugs (NSAIDs) can be used in more severe cases [8, 9]. Surgical intervention can be considered if the clinical response is unsatisfactory and all nonsurgical treatment options have been exhausted. Several surgical techniques have been described to correct these deformities, including soft tissue and bony procedures, with varying results [10,11,12,13].

As reported by Evans [14] and modified by Mosca [15], lateral column lengthening remains the most reliable option in children as it does not involve fusion, and it corrects both hindfoot and midfoot deformities [16, 17].

However, Evan’s procedure does not address the problems of the medial column, including forefoot supination. Therefore, Evan’s osteotomy is commonly combined with medial column procedures such as Young tenosuspension, tibialis posterior repositioning, tibialis anterior tendon transfer, or talonavicular joint capsular plication [10, 18,19,20].

One of the procedures that address the medial arch problems is tibialis anterior tendon plantar rerouting to serve as a plantar muscle and as a support to the talar head [21].

The current study aimed to assess the clinical and radiological results of single-stage reconstruction of symptomatic flexible flatfoot in pediatrics and adolescents using lateral column lengthening osteotomy and tibialis anterior tendon plantar rerouting.

Methods

This study was a prospective case series of children and adolescents with symptomatic flexible flatfoot treated by single-stage reconstruction using rerouting of the tibialis anterior tendon combined with calcaneal lengthening osteotomy from January 2014 to April 2016. A single pediatric orthopedic surgeon operated on all patients at the Pediatric Unit of the Orthopedic Surgery Department, Menoufia University Faculty of Medicine, after approval of the Institutional Review Board (IRB) and written consent of all patients’ parents.

Inclusion criteria were children and adolescents between 9 to 15 years with idiopathic symptomatic flexible flatfoot that failed to respond to conservative management for at least 3 months, no previous reconstructive surgery for the affected foot, and no local or general contraindications for surgery. Patients had to complete at least 2 years of follow-up to be included in this investigation. Rigid flatfoot and patients with associated neuromuscular disorders or syndromes were excluded.

Preoperative evaluation

Patients were evaluated preoperatively, both generally, locally, and radiologically. Generally, to assess body build, fitness for surgery, and associated neuromuscular disorders or syndromes. Locally, to evaluate the type of flatfoot, whether flexible or rigid using the Hubscher maneuver (Jack’s test) and the soft tissue condition of the foot.

Radiologically, standing anteroposterior (AP) and lateral foot X-rays were obtained to measure the preoperative parameters, including the standing talo-first metatarsal angle and talonavicular coverage angle in the AP view and talo-first metatarsal angle and calcaneal pitch angle in the lateral view.

In bilateral cases, we operated each side separately with a 4 to 6-month interval.

Surgical technique

Surgeries were done with patients in the supine position under general anesthesia. Antibiotic prophylaxis, third generation cephalosporins, was used two hours before the time of anesthesia induction at a dose of 50 mg/kg. A pneumatic tourniquet was used in all patients.

Calcaneal lengthening osteotomy

A 5 to 6 cm lateral longitudinal incision was centered over the calcaneocuboid joint. The sural and superficial branches of the peroneal nerve were protected. Dissection took place between the extensor digitorum brevis dorsally and the peroneal muscles plantarly. The calcaneocuboid joint (CCJ) was determined without joint capsule opening. The CCJ was fixed using 1.5 mm K-wire inserted from distal to proximal and stopped before the planned osteotomy site to prevent subluxation of the CCJ while opening the osteotomy site. Z-lengthening of the peroneus brevis was performed. Localization of the osteotomy site was done by the insertion of a reference K-wire about 1–1.5 cm proximal to the calcaneocuboid joint and parallel to it. The osteotomy was done between the anterior and middle facet of the calcaneus using an oscillating saw just proximal to the reference wire and under C-arm control. The osteotomy site was then opened using a lamina spreader up to the maximum opening to determine the width of the graft needed. A tricortical trapezoidal-shaped bone graft was taken from the ipsilateral iliac crest. The width of the graft ranged from 8 to 12 mm at one edge and 4 to 6 mm on the opposite edge. The graft was inserted and well-fitted into the osteotomy site. Although the graft was inherently stable, additional stability was added using a single staple to prevent dislodgement of the graft, Fig. 1.

Fig. 1
figure 1

Surgical steps on the lateral side. A Stabilization of the calcaneocuboid joint and insertion of a reference wire for the osteotomy site. B Osteotomy of the calcaneus using the oscillating saw. C A harvested trapezoidal tricortical iliac graft. D Application of the graft and stabilization with a stable

Full size image

Tibialis anterior rerouting

An 8 to 10 cm incision was made just medial to the tibialis anterior tendon from the first metatarsal base distally to the talonavicular joint proximally. The incision was deepened, and the tibialis anterior tendon was freed up along its route, keeping it attached to its insertion. A U-shaped osteoperiosteal flap was made on the medial side of foot, extending over the capsule and overlying the talonavicular joint, navicular bone, navicular-medial cuneiform joint and medial cuneiform bone keeping the flab attached at the planter side. A gutter was then created beneath the navicular tuberosity, and the tibialis anterior tendon was then pulled down into this gutter and stitched to soft tissues beneath the talar head. Two K-wires were inserted into the navicular as anchors to prevent slippage of the tendon back till soft tissue healing. The U-shaped osteoperiosteal flap was then sutured back into its place over the tibialis anterior tendon keeping the tendon in the gutter and preventing it from dislocating, Fig. 2.

Fig. 2
figure 2

Tibialis anterior rerouting. A Medial skin incision. B Releasing the tendon. C Pulling the tendon to the planter surface of the navicular. D Keeping the tendon in place with sutures. E K-wires insertion into the navicular to keep the tendon in place. F Final corrected foot shape

Full size image

Fractional lengthening of the gastrocnemius muscle

After completion of calcaneal lengthening and tibialis anterior rerouting, testing for ankle dorsiflexion was done. The ankle should be dorsiflexed to 10–15 degrees. In cases with ankle contracture, the Silfverskiold test was used to evaluate the cause of the contracture, whether it was gastrocnemius muscle or the Achilles tendon. Patients with positive Silfverskiold test underwent fractional lengthening of the gastrocnemius muscle through a separate posterior approach, Fig. 3.

Fig. 3
figure 3

Fractional lengthening of the gastrocnemius muscle. A Posterior skin incision. B Fractional lengthening of the tendinous part of the gastrocnemius muscle at the musculotendinous junction

Full size image

After completion of the surgical procedures, the tourniquet was deflated, and proper hemostasis was done before wound closure and application of a below well-padded knee cast.

Postoperative care

Immediate postoperative foot X-rays were done to assess the degree of correction of the radiological parameters. Postoperative antibiotic prophylaxis was given for 3 days in the form of third generation cephalosporins with a dose of 50 mg/kg/day. Patients were discharged from the hospital on the second day postoperative. Follow-up was done in the outpatient clinic two weeks postoperatively for wound assessment, stitches removal, and application of a new below-knee cast. Six weeks later, the cast was removed, and new foot X-rays were done to assess the healing of the calcaneal osteotomy. Below-knee walking cast was then applied and partial weight bearing was allowed for 4 weeks. This was followed by cast removal and full weight-bearing. Regular follow-up visits were advised every 6 months.

The American Orthopaedic Foot and Ankle Society (AOFAS) score [22] was utilized to assess the functional results of patients. Results were excellent with a score of 85 or more, good with a score of 75–84, fair when the score was 60–74, and bad when the score was less than 60.

Statistical analysis

Data were analyzed using IBM SPSS software package version 20.0. (Armonk, NY: IBM Corp). Continuous variables were represented as mean ± SD. Qualitative data were represented as numbers and percentages. Wilcoxon signed-rank test was utilized to compare quantitative variables which were not normally distributed. Results were judged significant at a 5% level.

Results

Demographics and baseline characteristics

This study included 28 feet in 16 patients, with a mean age of 11.6 ± 2.1 (range, 9–14.5) years. Eight (50%) patients were males and eight (50%) were females. The affected side was bilateral in 12 (75%) patients, right in 3 (18.8%) patients, and left in 1 (6.2%) patient. The main complaint was pain in 13 (46.4%) feet, cosmetic issues in 3 (10.7%) feet, and pain and cosmetic issues in 12 (42.9%) feet.

The average preoperative AOFAS score was 51.6 ± 5.5 (range, 42–61).

In 25 (89.3%) feet, there was ankle contracture with a positive Silfverskiold test that necessitated fractional lengthening of the gastrocnemius muscle through a separate posterior approach. The remaining 3 (10.7%) feet had no ankle contracture, Table 1.

Table 1 Demographics and baseline characteristics of the included patients
Full size table

Functional and radiological outcomes

The average follow-up duration was 38.9 ± 10 (range, 24–60) months. The average AOFAS score increased from 51.6 ± 5.5 (range, 42–61) preoperatively to 85.3 ± 10.2 (range, 56–96) at final follow-up, P < 0.001. Excellent results were obtained in 17 (60.7%) feet, good results in 8 (28.6%) feet, fair results in 1 (3.6%), and poor results in 2 (7.1%) feet. The 3 cases with fair and poor results complained of some degrees of pain, limited recreational activities, and hindfoot motion restriction. Achilles tendon lengthening was not performed in those 3 feet.

The talonavicular coverage angle decreased from 13.6 ± 4.4° (range, 5–25) preoperatively to 3.9 ± 3° (range, 0–10) postoperatively, P < 0.001. The AP talo-first metatarsal angle reduced from 16.9 ± 4.4° (range, 10–25) preoperatively to 4.5 ± 3.6° (range, 0–15) postoperatively, P < 0.001. The lateral talo-first metatarsal angle decreased from 19.2 ± 4.9° (range, 10–30) preoperatively to 4.6 ± 3.2° (range, 0–12) postoperatively, P < 0.001. The calcaneal pitch angle increased from 9.6 ± 1.9° (range, 5–13) preoperatively to 23.8 ± 4.8° (range, 12–30) postoperatively, P < 0.001, Table 2, Fig. 4.

Table 2 Preoperative and postoperative radiological parameters
Full size table
Fig. 4
figure 4

AP and lateral X-ray images at the weight-bearing position of a 12-year-old female child. A Preoperative AP talo-first metatarsal angle (17°) and talonavicular coverage angle (15°). B Preoperative lateral talo-first metatarsal angle (27°) and calcaneal pitch angle (10°). C Postoperative AP talo-first metatarsal angle (5°) and talonavicular coverage angle (7°). D Postoperative lateral talo-first metatarsal angle (5°) and calcaneal pitch angle (24°)

Full size image

Complications

No major complications developed. Three feet got superficial wound infection which was managed by dressing and antibiotics according to culture and sensitivity. Five feet developed hardware irritation from the medial K-wires that were removed after a sufficient period for complete soft tissue healing (4–6 months postoperatively).

Discussion

Flexible flatfoot is a frequent three-dimensional deformity in children and adolescents which can be asymptomatic or present with pain and gait disturbances [23]. The treatment aims to relieve the symptoms and to realign the foot [24]. Corrective surgery is sometimes necessary for moderate or severe deformities with failed conservative measures [10, 20, 25]. Joint arthrodesis is not advised for correction of flexible flatfoot as it can lead to stress shifting to adjacent joints and early arthritic changes [1].

As a result of their unreliable results and deformity recurrence, isolated soft tissue procedures are no longer advocated [8]. Combined bony and soft tissue procedures are most effective in correcting the deformity in all three planes [1, 20, 21, 24, 26]. By correcting the deformity, the foot kinematics improve during weight-bearing, which reduces the rapid progression of the deformity, which could cause degenerative changes to the foot and the need for future arthrodesis [16, 27].

Evans lateral calcaneal lengthening osteotomy can address the collapsed medial arch and heel eversion, but this osteotomy alone ignores other elements of the flatfoot deformity [28, 29]. Several medial column soft tissue procedures have been reported, such as tibialis anterior transfer to the navicular and tibialis anterior or posterior tenodesis [30].

In our study, a reconstructive technique of tibialis anterior rerouting combined with calcaneal lengthening osteotomy was used. A trapezoidal iliac graft was used in the calcaneal osteotomy to keep the lengthening, as advised by Mosca [15]. Despite the graft being inherently stable after application, additional fixation with a stable was done to save guard against dislodgement of the graft, which had not occurred in any case.

Kissel and Blacklidge [31] transferred the tibialis anterior tendon via the talus from lateral dorsal to medial plantar, with suturing of the remaining tendon to the soft tissues.

In our study, the distal 10–15 cm of the tibialis anterior tendon was mobilized and redirected in an L course without disruption of the tendon insertion. The horizontal limb acted as a strong plantar ligament, and the vertical limb was stitched to the undersurface of talonavicular joint to support the subluxated joint. After its rerouting, the tibialis anterior can still act as an ankle dorsiflexor, as it mainly functions during the swing phase [32].

In this study, Achilles tendon lengthening was done in 25 feet through fractional lengthening of the musculotendinous junction. Achilles tendon contracture often coexists with symptomatic flatfoot, preventing normal ankle dorsiflexion during midstance [33]. In these cases, Achilles tendon lengthening is necessary to allow the ankle to dorsiflex at least 10–15 degrees [34, 35]. In our study, Achilles tendon lengthening was not performed in 3 cases that had fair and poor results. Therefore, we advise that ankle equinus should be thoroughly examined after correction of the deformity for lengthening Achilles tendon or Gastrocnemius fractional lengthening in positive Silfverskiold test.

In our study, the average AOFAS score increased significantly from 51.6 ± 5.5 preoperatively to 85.3 ± 10.2 at the last follow-up. Our results were comparable to previous studies that reported other reconstructive methods. Oh et al. [36] reported the outcomes of combined reconstruction of the symptomatic flexible flatfoot in adolescents using medializing calcaneal osteotomy and lateral column lengthening. The average AOFAS score increased significantly from 49.1 preoperatively to 93.4 at last follow-up [36]. Viegas et al. [30] reported significant improvement of the average AOFAS score from 68.59 preoperatively to 96.35, following Evans calcaneal osteotomy combined with medial split tibialis anterior tendon transfer, in 34 flexible planovalgus feet, with an average follow-up of 24.9 months.

In our study, all the radiological parameters improved significantly postoperatively, including the AP and lateral talo-first metatarsal angle, talonavicular coverage angle and calcaneal pitch angle. Similarly, Oh et al. [36] and Viegas et al. [30] reported improvement in those radiological procedures following their technique.

Lateral calcaneal lengthening osteotomy must be performed with caution. Li et al. [37] conducted a cadaveric study on the optimal size of graft size and reported that a 6 mm graft was not associated with any significant calcaneal shift, while grafts of 8 mm or more led to significant anterior and posterior shift of two calcaneal segments and posterior facet impingement.

The idea of tibialis anterior rerouting resembles Young’s tenosuspension procedure [38], which reroutes the tibialis anterior tendon through a navicular key-hole slot. In our study, tibialis anterior rerouting was performed without separating the tendon from its insertion. The tibialis anterior tendon was not rerouted into the talar head, but it was rerouted plantarly beneath the navicular bone to support the talonavicular joint, acting as a second spring ligament to support the medial arch. As the distal part of the tibialis anterior tendon was left attached to its insertion, rerouting of the tendon plantarly increased its tension and thus creating a supination moment of the foot without affection of its function as a dorsiflexor of the tibiotalar joint during the swing phase and act as a static stabilizer of the talonavicular joint preventing the dropping of the medial arch during the stance phase.

The rerouting of the tibialis anterior tendon beneath the navicular, however, may increase forefoot supination postoperatively. This soft tissue procedure weakens the dorsiflexion power of the tibialis anterior, leading to compensatory overactivity of toe extensors and a mechanical advantage to the plantar flexion power of the peroneus longus that might evolve over time [39].

The tibialis anterior tendon mainly functions during the swing phase, not the stance phase [32]. Therefore, it is questionable whether this procedure has an active effect on the rise of the medial arch, which is more likely done by the unantagonized long peroneal function. To investigate the possible long-term drawbacks and biomechanical changes of this procedure, a long-term follow-up study is needed.

The procedure presented in this study was adopted from El-Tayeby study [21], that reported treatment of severe flexible flatfoot with combined Evans calcaneal osteotomy and tibialis anterior rerouting in addition to navicular-first cuneiform wedge resection and arthrodesis for shortening of the medial column of the foot. We did not perform naviculocuneiform arthrodesis as this additional invasive bony procedure may result in limitation of foot motions, overcorrection of the deformity, and unexpected long-term outcomes.

This study has some limitations, such as the relatively low number of operated feet and the relatively non-adequate follow-up period that is not enough to address the long-term outcomes of this procedure on foot, such as degenerative arthrosis. We did not perform footprint analysis by a pedograph as it is costly and not widely available in our country enough to be performed on a regular basis. The absence of comparison between the results of this procedure and other procedures is another limitation of the study. However, the combined technique described in this study is a good option for addressing and maintaining all components of flatfoot deformity.

Conclusion

Surgical reconstruction of symptomatic flexible flatfoot with lateral column lengthening and tibialis anterior rerouting has good radiological and clinical outcomes with a considerably low complication rate.