Secondary Logo

Journal Logo

Special Focus

The Collapsing Flatfoot: Bone Alignment, Bone Alignment, Bone Alignment!

Roberts, Lauren E. MD, MSc, FRCSC*; Ellis, Scott J. MD

Author Information
Techniques in Foot & Ankle Surgery: December 2019 - Volume 18 - Issue 4 - p 185-193
doi: 10.1097/BTF.0000000000000238
  • Free



Although the medializing calcaneal osteotomy (MCO) is one of the most common procedures performed for a patient with adult acquired flatfoot deformity in North America, there has historically been limited evidence guiding the exact amount of correction to be performed. Work done previously at our institution has shown that a patient’s overall outcome is most closely associated with the correction of hindfoot alignment through the MCO above all other factors and that this can be modeled linearly.1 The exact amount of correction required does differ with each patient as each patient presents with a different amount of hindfoot deformity and may respond differently to different amounts of calcaneal medialization. Specific recommendations to slide the heel 10 mm medially have been made in various technique papers in the past,2–5 however, only 1 biomechanical study supports this amount of correction.6 Overall, there is limited clinical evidence to support this one size fits all approach.7 We therefore advocate for determining preoperatively a level of correction that is specific to each patient’s preoperative deformity.

When forefoot abduction is present, a lateral column lengthening (LCL) procedure is often also required. Lengthening of the lateral column is performed in this setting to improve talar head coverage by the navicular and elevate the medial longitudinal arch while also improving the forefoot abduction.4,5,8 The Evans procedure is a commonly used procedure that has been successful in the literature in terms of realignment, however, it does have some potential complications to consider such as nonunion, lateral column overload, fifth metatarsal stress fractures, and stiffness.9 An issue with early literature is that there has historically been no measure of exactly how much LCL is required. Overlengthening of the lateral column is often the cause of the aforementioned complications with this technique.

The stepcut lengthening calcaneal osteotomy (SLCO) as described by Griend10 is an alternative technique for lengthening the lateral column. Theoretical benefits of this technique include a lower risk of malunion as the position of the longitudinal arm of the stepcut osteotomy is not dependent on any interposed graft. Overall, this potentially results in a lower risk of lateral foot overload due to the decreased risk of elevation malunion of the distal segment.4,8,10


When conservative measures fail, surgical correction can be considered. Before indicating a patient for a bony realignment procedure through osteotomies, the absence of significant symptomatic hindfoot joint arthritis must be confirmed as these may require fusion procedures. In the case of a true stage II flatfoot, the deformity is passively correctable and there is no evidence of any significant arthritis. In these patients, osteotomies for bony realignment along with soft tissue procedures are indicated. In our experience, patients with lateral subluxation of the subtalar joint at the level of the medial facet, sinus tarsi impingement or subfibular impingement may in fact be better treated with subtalar fusion rather than osteotomy.

Bony procedures for realignment aim to address 3 components; hindfoot valgus, forefoot abduction, and residual forefoot supination. In this chapter, we will discuss our approach to achieving optimal hindfoot realignment specifically through a MCO and further achieving appropriate midfoot coronal alignment through LCL procedures such as the Evans osteotomy or the SLCO. Techniques and considerations for correction of residual forefoot supination are discussed elsewhere.


Once a symptomatic patient is deemed suitable for correction through osteotomies rather than fusion, various measures to assist with calculating the amount of correction and specific procedures to be used can be considered (Figs. 1, 2). That said, the specifics on the amount and approach to realignment must be definitively decided intraoperatively. Hindfoot moment arm as measured on hindfoot alignment radiographs is shown in Figure 1. Two points are marked on the tibia at 10 and 15 cm above the plafond. An axial line intersecting these 2 points is created. The hindfoot moment arm is the shortest distance from this midaxial tibial axis to the inferior most aspect of the calcaneus. Previous work has shown that the degree of intraoperative calcaneal translation correlates with the change in preoperative and postoperative hindfoot moment arms by a specific equation detailed by Chan et al.1

Hindfoot alignment measurement. A, Preoperative image demonstrating a hindfoot moment arm of 18.5 mm. B, Postcorrection measurement demonstrating a neutral hindfoot moment arm.
Forefoot abduction measurement. Preoperative anteroposterior radiographs of the foot demonstrate the various preoperative parameters of foot abduction. The lateral incongruency angle (A), the talar uncoverage angle (B), and the first talar metatarsal angle (C).

Preoperative anteroposterior radiographs of the foot should then be reviewed to define the amount of forefoot abduction. The lateral incongruency angle, the talar uncoverage (TUC) angle, and the first talar metatarsal (TMT) angle (Fig. 2) are the 3 parameters we assess in a preoperative patient. The lateral incongruency angle (Fig. 2A) is formed by the intersection of a line from the lateral talar articular surface (b′) and the lateral most navicular articular surface (c′) with a line formed between b′ and the narrowest aspect of the talar neck (a′). The TUC (Fig. 2B), is measured between the lines connecting the medial and lateral most points on the articular surface of the talus and navicular respectively. The first TMT angle is the angle between the longitudinal axis of the first metatarsal and the talus respectively (Fig. 2C). Previous work has shown that the amount of LCL performed correlates with the change in in the lateral congruency angle again based on a specific equation again detailed by Chan et al.11


We typically perform a flatfoot reconstruction under a spinal block combined with a popliteal fossa block and conscious sedation. The patient is positioned supine with a wedge pillow beneath the operative limb to optimize ease of obtaining fluoroscopic views intraoperatively (Fig. 3). An above knee tourniquet is placed. We preferentially use allograft wedges with iliac crest bone marrow aspirate concentrate (BMAC) to avoid the donor site morbidity and potential complications of autograft harvest. For that reason, iliac crest bone marrow aspirate can be obtained before prepping out the leg in the usual manner (Fig. 3D).

Patient positioning. The patient is positioned supine with a wedge pillow beneath the leg (A) and a bump beneath the ipsilateral hip (B). The option for external rotation should be maintained (C). Iliac crest aspirate can be before (D) or after draping.

All anticipated potential incisions are then marked as shown in Figure 4. Our typical approach involves first performing a gastrocnemius recession if indicated and then assessing and treating the medial soft tissue structures through an incision over the posterior tibial tendon (Fig. 4A). Attention is then drawn to bony realignment procedures based on the preoperative plan. The incision for the MCO lies obliquely between the “soft spot” of the retrocalcaneal bursa and the “soft spot” anterior to the calcaneal tuberosity distally (Fig. 4C). The second lateral incision for a potential LCL is similar to the standard sinus tarsi approach but can be made slightly more plantar to allow for ease of plantar dissection as long as the skin bridge is respected.

Surgical incision planning. Incisions are marked preoperatively for the gastrocnemius recession and medial soft tissue work (A), the cotton osteotomy (B), and the medializing calcaneal osteotomy and lateral column lengthening (C). Ensuring adequate skin bridges is essential.

The posterior most incision is then made sharply through the skin taking care to avoid the sural nerve and then sharply down to bone. Electrocautery is then used to mark an oblique line on the lateral wall of the calcaneus in line with the skin incision. A periosteal elevator is then used to elevate the periosteum both dorsally and plantarly from this point. A self-retaining retractor is then placed centrally with a Hohmann retractor placed around the tuberosity at each apex of the incision. To plan for the MCO cut, 2 Kirschner (K)-wires are first placed in line with the skin incision, at an approximate angle of 45 degrees from the plantar aspect of the foot (Fig. 5). A combination of saw and osteotome is then used to complete the osteotomy.

Planning the medializing calcaneal osteotomy. Two pins are placed in line with the skin incision to plan for the osteotomy (A). This oblique osteotomy should exit proximally at the most inferior aspect of the slope of the calcaneal tuberosity as marked by the arrow (B).

A blunt laminar spreader is then placed within the osteotomy to ensure it is fully released. Any remaining restrictive bands limiting mobility are sharply released. The laminar spreader is then removed and the surgeon ensures good mobility of the tuberosity. Two K-wires are placed through the heel but not yet across the osteotomy site (Fig. 5A) so that their exit point within the osteotomy can be observed through the lateral incision to ensure an appropriate trajectory. The surgeon then opposes the osteotomy, translates the heel medially and passes the K-wires across the osteotomy (Fig. 6A). The amount of translation is then measured (Fig. 6B). The corrected hindfoot alignment is then assessed as shown in Figure 6C.

Medializing calcaneal osteotomy. Kirschner-wires are placed centrally in the heel into the osteotomy (A). The tuberosity is then translated and pinned. Neutral clinical heel alignment is confirmed (B). Translation is then measured (C) and maintenance of eversion is confirmed (D).

Most importantly, once the MCO is pinned, maintenance of subtalar joint range of motion in eversion is confirmed (Fig. 6D) clinically. The heel will then be assessed using lateral fluoroscopy to ensure appropriate contact at the osteotomy site, spread of the K-wires and neutral to slight inferior placement of the tuberosity to help reconstitute the calcaneal pitch (Fig. 7). A Harris heel view is then used to ensure appropriate bony contact in the coronal plane and safe placement of the wires (Fig. 7B). The MCO is then fixed with two 4.5-mm cannulated screws at this stage before moving on to the LCL. Final images postfixation are shown in Figure 7C.

Fixation of the medializing calcaneal osteotomy. After positioning of the heel, the final degree of medial translation is shown here (A). The resultant ledge seen here is later removed with a combination of the saw or rongeur. Once fluoroscopic position of the heel and wires is confirmed (B), final 4.5-mm partially threaded cannulated screws are placed (C).


After correcting the heel valgus with a calcaneal osteotomy, if there is still residual forefoot abduction, LCL is indicated. As previously outlined, there are 2 primary techniques that can be used: an Evans osteotomy or a SCLO. Both of these osteotomies result in an improvement to forefoot abduction, each with their own considerations. In general, although more technically challenging, the SLCO may have better rates of union. Differences between these 2 osteotomies otherwise are limited and indications to proceed with one over the other therefore often relates to surgeon preference.

Most typically we next proceed with a LCL through the anterior calcaneus, originally described by Evans.12 Classically, this technique involved a lateral opening wedge of 1 cm, made 1.5 cm proximal to the calcaneocuboid joint. Similar to Hintermann et al,13 however, we favor a placement of this osteotomy 12 mm from the calcaneocuboid joint to avoid disrupting any articular surface. The previously marked anterior, most lateral incision is made sharply through skin (Fig. 4C). The peroneus brevis and longus tendons are protected inferiorly. The extensor digitorum brevis is elevated off the anterior calcaneus superiorly and further subperiosteal elevation is performed plantarly until the inferior border of the calcaneal neck is reached. A Hohmann retractor is placed both plantarly and dorsally over the calcaneus at a distance of 12 mm from the calcaneocuboid joint. A K-wire is then used to plan the starting point and trajectory of the osteotomy as seen in Figures 8A, B.

Planning for an Evans lateral column lengthening. A Kirschner-wire 12 mm from the joint for the planned osteotomy is confirmed clinically (A) and on fluoroscopy (B). The wire is then cut and serves as a guide for the saw (C).

The Evan’s osteotomy is performed in line with this wire and taken to, but not through the medial cortex. A combination of saw and osteotome can then be used to ensure this osteotomy is complete both plantarly and dorsally. A self-retaining pin distractor is then placed laterally spanning the osteotomy with a K-wire placed on either side as seen in Figure 9. The osteotomy is opened with a self-maintaining distractor to the desired distance calculated preoperatively. The resultant incongruency angle and the amount of talar head coverage is then assessed radiographically and residual eversion is assessed clinically (Figs. 9B, C).

Determining the amount of lateral column lengthening required. After completion of the osteotomy, a self-maintained pinned distractor is used to open the osteotomy site (A). This is checked fluoroscopically to assess the incongruency angle and the degree of talar head coverage (B). This retractor holds the osteotomy open while the surgeon assesses the foot clinically for residual forefoot abduction and residual eversion motion (C).

The Evan’s LCL is then measured using a ruler and an allograft wedge of this size is then prepared and soaked in BMAC to aid in healing. The graft is then placed within the osteotomy and once appropriately aligned, the distractor is released, compressing the graft wedge (Fig. 10A). This is provisionally held in place with a 0.62 K-wire from distal to proximal as definitive fixation is then performed. We preferentially utilize a lateral locking compression plate (Fig. 10C) but a single screw or other constructs could also be used instead.

Fixation of the Evans lateral column lengthening. A bone graft wedge is prepared and placed after adequate opening is confirmed (A). It is pinned (B) to allow for clinical reevaluation by the surgeon and fixation of the graft. In this case, a 2-hole compression plate was used (C).

Alternatively, the SLCO is performed through a similar incision centered over the anterior process of the calcaneus. Dissection posteriorly extends to the level of the peroneal tubercle and the peroneal tendons are identified. The sheath of the peroneal tendons is incised, the tendons are released and retracted and protected inferiorly (Fig. 11B). Two 1.6-mm K-wires are then placed within the calcaneus at the corners of the stepcut to plan the osteotomy and confirm these cuts on fluoroscopy before proceeding. The distal vertical cut should be ∼10 to 12 mm from the coracoclavicular joint, similar to the Evans osteotomy. The horizontal limb of the stepcut is extended from this point posteriorly to a point just distal to the peroneal tubercle. This limb is made as long as possible, but care must be taken to ensure it does not extend into the zone of the MCO. The proximal vertical limb is then extended plantarly from here. Placement of these K-wires are checked on fluoroscopy to ensure appropriate distance from the coracoclavicular joint and the posterior calcaneal osteotomy.

The stepcut calcaneal osteotomy. The stepcut lengthening calcaneal osteotomy is performed through a lateral incision overlying the anterior process of the calcaneus ensuring a maximized skin bridge (A). The peroneals are retracted plantarly (B). Once the osteotomy cut is made, a self-maintained pin distractor is placed to distract and rotate through the osteotomy (C).

A microsagittal saw is then used to complete all 3 limbs of the osteotomy taking care to avoid the medial soft tissue structures. The same saw is then used in an oblique fashion to shave the medial cortex on both vertical limbs to ensure there will be no impingement when correction is made. A distractor is then placed to distract and rotate through the osteotomy to obtain adduction of the forefoot (Fig. 11C). The size of the vertical limb gaps are measured using a ruler and appropriate sized wedges are fashioned using allograft (with BMAC) or autograft. The wedges are then placed laterally in the vertical gaps created by the osteotomy and provisionally pinned. Definitive fixation is then achieved with either 2.7- or 3.5-mm cortical screws as shown in Figure 12. Healing typically occurs along this horizontal limb rather than at the allograft sites therefore typically speeding the time to functional healing and decreasing the rate of nonunion associated with allograft.

Postoperative radiographs after stepcut lengthening calcaneal osteotomy. Final anteroposterior (A) and lateral (B) radiographs after a medializing calcaneal osteotomy and a stepcut lengthening calcaneal osteotomy fixed with two 3.5-mm screws and one 2.7-mm screw.


At our institution we assessed 55 patients undergoing reconstruction for adult acquired flatfoot deformity without hindfoot fusions. Foot and Ankle Outcome Score, a legacy functional outcome tool previously validated for flatfoot patients,14 were obtained preoperatively and >22 months postoperatively.7 The hindfoot alignment of each patient was assessed with a hindfoot alignment radiograph (Fig. 1) >48 weeks postoperatively. Patients were categorized as having either mild varus, moderate varus or valgus. Patients with a mild radiographic overcorrection (0 to 5 mm of varus) had the best clinical outcome.7 In our experience, this mild radiographic overcorrection often correlates with a clinically neutral hindfoot on the operating table. This scenario is correlated with the best results for hindfoot valgus correction.

With regards to LCL, a study by Saunders et al15 reviewed the results of 111 patients who underwent flatfoot reconstruction and compared the outcomes between the Evans LCL and the SLCO. In their results, patients treated with a SLCO procedure had faster healing times and fewer nonunions than those who had an Evans LCL. The Foot and Ankle Outcome Score and degree of correction were not significantly different between the 2 techniques.15

A study at our institution was also performed looking at the correlation between residual forefoot position after Evans osteotomy and patient outcomes.16 Fifty-five patients were classified as either having forefoot abduction or adduction postoperatively based on the lateral incongruency angle, TUC angle and the Talo-first metatarsal angle (Fig. 2). In this study, abduction was defined as 2 of the following: lateral incongruency angle of >5 degrees, a first TMT angle of >8 degrees and/or a TUC angle of >8 degrees.16 Analysis of these patients demonstrated that patients with slight abduction or normal midfoot position had the best outcome scores. Patients overcorrected into adduction had worse outcomes.16 It is our opinion that as long as the appropriate amount of LCL is performed based on the measures noted above, in a patient with no risk factors for nonunion, the Evans type osteotomy is the preferred technique for LCL. The Evans procedure is technically more straightforward, involves less dissection around the peroneal tendons and often requires less time to perform than the SCLO while providing similar outcome scores and correction.15

Although the MCO is known the be the most important determinant of hindfoot valgus, the LCL does offer a triplane correction, so may limit somewhat the degree of MCO that is required. Some surgeons will pin each correction provisionally and then perform final fixation when satisfied with the overall, combined appearance of the foot alignment intraoperatively. We believe that because the MCO is the most important step for correcting hindfoot valgus, it should be performed and fixed first before proceeding to the LCL.


As previously discussed, an important potential complication to consider with both the LCL and MCO is overcorrection. Overcorrection can lead to varus through the hindfoot or loss of eversion if the forefoot abduction is overcorrected. If the patient is too stiff in eversion, there are cases when the amount of MCO performed should be decreased or the LCL not performed at all. Leaving a patient stiff in eversion will lead to lateral border overload which is not well tolerated.17 This can also lead to metatarsal stress fractures of the fifth metatarsal. In our experience, there are also cases when the amount of translation calculated preoperatively is too great and may limit eversion. In these cases, the preoperatively calculated amount of translation should be used only as a rough guide.

The more common complication in our practice is painful hardware. The heel screws we typically use to fix the MCO are headed, and although we ensure they are well seated, they can still become symptomatic in some cases. Only symptomatic screws are removed and only after a minimum 1 year after surgery. Nonunion of the LCL is also a potential complication. The rate of nonunion for LCLs at our institution is 10.3%. Loss of correction through the LCL is also possible at a rate of 5.6%.9 A study at our institution compared outcomes for LCL procedures with autograft versus allograft and found that complication rates and rates of union were not significantly different between the 2 types of graft.9 Patients who present with pain and swelling over the lateral foot and have not healed by 6 months are candidates for revision of their LCL nonunion. Other reasons for revision of the LCL include overcorrection as outlined above and undercorrection or a prominent graft causing impingement.


Postoperatively, the patient is placed in a well-padded posterior plaster splint. Typically, these patients remain non–weight-bearing for 6 weeks if an LCL was not needed. If they underwent a LCL, then they are non–weight-bearing for 8 weeks postoperatively. At 2 weeks, sutures are removed and they are transitioned to a CAM walker boot. Clinical and radiographic reassessment next occurs at 6 weeks at which point they progress to weight-bearing as tolerated over the next 4 to 6 weeks in the CAM walker boot. The patient is seen again at 3 months at which point they transition into a sneaker.


We continue to learn about new features of the stage II flatfoot that should be considered when developing a surgical plan. Patients with subfibular impingement, sinus tarsi impingement, and middle facet subluxation will need to be better understood to determine if their outcomes are best with treatment through osteotomies versus selective hindfoot fusion. Potentially modifiable concerns with these procedures include rates of union and wound problems. Increasing use of the minimally invasive cutting burr techniques could have an application for the MCO in this setting especially when dual posterolateral incisions would otherwise be used as a narrow skin bridge is often encountered. Utilizing a minimally invasive cutting burr here may help address this. Growing evidence on biologic agents that may improve healing rates when combined with allograft may improve outcomes moving forward as well.


1. Chan JY, Williams BR, Nair P, et al. The contribution of medializing calcaneal osteotomy on hindfoot alignment in the reconstruction of the stage II adult acquired flatfoot deformity. Foot Ankle Int. 2013;34:159–166.
2. Mosier-LaClair S, Pomeroy G, Manoli A II. Operative treatment of the difficult stage 2 adult acquired flatfoot deformity. Foot Ankle Clin. 2001;6:95–119.
3. Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon deficiency with flexor digitorum longus tendon transfer and calcaneal osteotomy. Foot Ankle Int. 2004;25:445–450.
4. Haddad SL, Myerson MS, Younger A, et al. Symposium: adult acquired flatfoot deformity. Foot Ankle Int. 2011;32:95–111.
5. Guha AR, Perera AM. Calcaneal osteotomy in the treatment of adult acquired flatfoot deformity. Foot Ankle Clin. 2012;17:247–258.
6. Arangio GA, Salathe EP. A biomechanical analysis of posterior tibial tendon dysfunction, medial displacement calcaneal osteotomy and flexor digitorum longus transfer in adult acquired flat foot. Clin Biomech (Bristol, Avon). 2009;24:385–390.
7. Conti MS, Ellis SJ, Chan JY, et al. Optimal position of the heel following reconstruction of the stage ii adult-acquired flatfoot deformity. Foot Ankle Int. 2015;36:919–927.
8. Vulcano E, Deland JT, Ellis SJ. Approach and treatment of the adult acquired flatfoot deformity. Curr Rev Musculoskelet Med. 2013;6:294–303.
9. Vosseller JT, Ellis SJ, O’Malley MJ, et al. Autograft and allograft unite similarly in lateral column lengthening for adult acquired flatfoot deformity. HSS J. 2013;9:6–11.
10. Griend RV. Lateral column lengthening using a “Z” osteotomy of the calcaneus. Tech Foot Ankle Surg. 2008;7:257–263.
11. Chan JY, Greenfield ST, Soukup DS, et al. Contribution of lateral column lengthening to correction of forefoot abduction in stage IIb adult acquired flatfoot deformity reconstruction. Foot Ankle Int. 2015;36:1400–1411.
12. Evans D. Calcaneo-valgus deformity. J Bone Joint Surg Br. 1975;57:270–278.
13. Hintermann B, Valderrabano V, Kundert HP. Lengthening of the lateral column and reconstruction of the medial soft tissue for treatment of acquired flatfoot deformity associated with insufficiency of the posterior tibial tendon. Foot Ankle Int. 1999;20:622–629.
14. Mani SB, Brown HC, Nair P, et al. Validation of the Foot and Ankle Outcome Score in adult acquired flatfoot deformity. Foot Ankle Int. 2013;34:1140–1146.
15. Saunders SM, Ellis SJ, Demetracopoulos CA, et al. Comparative outcomes between step-cut lengthening calcaneal osteotomy vs traditional Evans osteotomy for stage iib adult-acquired flatfoot deformity. Foot Ankle Int. 2018;39:18–27.
16. Conti MS, Chan JY, Do HT, et al. Correlation of postoperative midfoot position with outcome following reconstruction of the stage II adult acquired flatfoot deformity. Foot Ankle Int. 2015;36:239–247.
17. Ellis SJ, Yu JC, Johnson AH, et al. Plantar pressures in patients with and without lateral foot pain after lateral column lengthening. J Bone Joint Surg Am. 2010;92:81–91.

medializing calcaneal osteotomy; hindfoot valgus; Evans osteotomy; stepcut calcaneal lengthening osteotomy; lateral column lengthening

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.