Insights on Management of Talar Avascular Necrosis

Avascular necrosis (AVN) of the talus, although rare in nature, provides considerable challenges to foot and ankle surgeons with management options contingent upon the severity of pathology and individualized functional outcomes for the patient. Owing to its unique structure, the talar surface is approximately 60 percent articular cartilage and provides no attachment sites for muscular or tendinous structures.¹ The vascular supply of the talus further follows a delicate pattern secondary to the extensive nature of the cartilaginous surface area, limiting available regions for perfusion.

Both traumatic and atraumatic causes exist for osteonecrosis of the talus. The etiology of talar AVN can further subdivide into anatomic location and mechanism of compromise. Extraosseous arterial compromise includes such phenomena as trauma, atheromata, and microemboli.² One may see intraosseous arterial obstruction with corticosteroid use, alcoholism, fat emboli, renal transplantation, sickle cell anemia, hyperlipidemia, irradiation, thrombophilias, and systemic disease processes including rheumatoid arthritis and systemic lupus erythematosus.^2,3

Clinicians may diagnose talar osteonecrosis through a combination of information gained from a comprehensive history and physical examination, and imaging studies. Although plain radiographs serve as the gold standard for initial screening of bony injuries, sensitivity for talar injury is only 74 percent, with displacement driving radiographic sensitivity.⁴ While computed tomography (CT) is more sensitive than radiographs in detecting signs of AVN, including marginated sclerosis, Hawkins, and crescent signs, magnetic resonance imaging (MRI) is ultimately the most sensitive imaging technique for detecting AVN in the early stages, present by the second week after infarction.⁵

To date, no level-one studies or consensuses exist to guide surgical management of talar osteonecrosis. Dhillon and team advocated for reserving nonoperative management for early diagnoses, intending to prevent collapse until completion of revascularization.⁶ The duration of non-weight-bearing (NWB) remains ill-defined in literature, with recommendations of NWB until fracture healing and completion of revascularization.⁶ Unfortunately, failure of nonoperative management frequently necessitates surgical intervention, which includes either joint-sparing or joint-sacrificing procedures. As evidenced in Hawkins’ series of nonoperative treatment, of their cohort, 58.3 percent later needed operative treatment, including 29.2 percent who underwent ankle or subtalar arthrodesis, 12.5 percent who underwent talectomies, and another 16.7 percent who underwent a bone grafting procedure.⁷

Considering joint-sparing treatments versus joint-sacrificing procedures, early surgical options when the talar body remains intact include core decompression and vascular bone grafting. However, once viability and collapse of the talar body occur, options include tibial calcaneal arthrodesis, the Blair form of tibial calcaneal arthrodesis, and utilization of allografts (fresh femoral head) or bioengineered talar implants. Future prospective, randomized studies are necessary to guide the conservative and surgical management of talar AVN; however, clinicians’ contemplation of treatment should be patient-centric based upon initial clinical presentation and progression of pathology.

When A Patient Presents for a Non-Healing Ankle Injury

A 54-year-old female without diabetes sustained a twisting injury while walking back to her car after a hockey game on some non-pavement grass surface and stepped into a hole. Consequently, she sustained a supination internal rotation left ankle injury. After X-rays and evaluation in the emergency room, the diagnosis was an ankle sprain. Treatment included a CAM boot with weight-bearing as tolerated. At a 3-week follow-up with her primary care physician, she needed crutches and could not bear full weight. Further X-ray evaluation did not identify any fractures.

She continued three more weeks of CAM boot immobilization and, at 6 weeks, was told to return to shoe wear and sent for physical therapy. She did not do physical therapy but tried to return to normal activities. She described continued pain, swelling, and stiffness of her left foot and ankle. Now, 12-weeks post-injury, she returned to her primary care physician, who ordered magnetic resonance imaging (MRI), revealing a Hawkins Stage 2 talar neck fracture. The fracture was partially healed, however, there were early signs of talar avascular necrosis. The patient received instructions to “get off it” and purchase a knee scooter. At 18-weeks post-injury, she sought another opinion, still unable to weight-bear. X-rays now showed a completely sclerotic talar body, consistent with advanced talar avascular necrosis. Repeat MRI showed articular collapse, but the body architecture remained intact. We then managed the patient surgically.

Due to complete loss of talar viability, we performed a talectomy with autograft to attain a tibial calcaneal arthrodesis. Her insurance company denied an attempt to acquire a bioengineered talus due to cost. We attempted a trans-fibular approach and used the resected fibula as an autograft. A 6.5 mm positioning screw helped gain length to the area of the resected talus. We also used fibular struts and crushed cortical cancellous autograft. Application of a 4.5 mm anatomic locking plate laterally to the tibia and calcaneus maintained correction and removed the stress load off the autograft. We noted good anatomic alignment and length on the postoperative X-rays and a plantigrade left foot and ankle.

Unfortunately, the patient began to note signs of infection at approximately 10 to 14 days. The wound was only mildly erythematous, but we noted copious serous drainage. The wound probed to the locking plate, and cultures grew methicillin-resistant Staphylococcus aureus (MRSA). We then removed the plate and screws as well as the autograft, which was also infected, and placed a vancomycin-loaded polymethylmethacrylate (PMMA) spacer at the excision site. A delta frame maintained stability to the foot and ankle soft tissue and the spacer. Closure took place over the spacer, and then we applied negative pressure wound therapy (NPWT) over the wound itself. This allowed management of any extensive serous drainage. The patient received a PICC line and 6 weeks of intravenous vancomycin.

Unfortunately, the patient still did not completely heal. The lateral wound continued to drain serous fluid in the wound vac containers but did not grow bacteria or show signs of inflammation. At this point, we recommended plastic surgical evaluation, resulting in a lateral rotational flap based on the peroneal artery, allowing closure of the wound. With this technique, the ankle itself remained unstable in its valgus state to allow soft tissue relaxation for the lateral flap. The process took nearly three months to heal completely. Radiographs showed biomechanical alignment at this time with a sidecar deformity secondary to loss of talar support.

Key Points to Understanding the Surgical Approach

At this point, we used computer-generated hexapod external fixation technology. We applied a circular frame to realign the ankle joint by soft tissue distraction utilizing olive pins in the tibia, calcaneus, and metatarsals. It took 6 weeks to attain a corrected position (see first and second photos to left). Also, complete restoration of the joint space allowed restoration of the talar body. We also noted that the distraction process reduced the bulk of the lateral flap.

After removing the frame, we performed a revision in that same surgical session. We utilized a posterior Gaille approach due to an uncompromised soft tissue conduit. Reflecting the Achilles tendon superiorly, we used an acetabular reamer to create space for the femoral head allograft. This allowed the removal of the soft tissue occupying the distracted joint space. A fresh frozen femoral head allograft underwent injection with bone marrow aspirate and contouring to fill the tibiocalcaneal void. We then utilized Augment Regenerative Solutions (Stryker/Wright Medical) at the contact points between the allograft and tibiocalcaneal surfaces (see third photo to left). A posterior anatomic locking plate and a 6.5mm headless positioning screw secured the graft (see fourth photo to left). A lateral radiograph exhibited excellent restoration of height (see fifth photo to left).

Again, misfortune occurred 10 days postoperatively with signs of wound dehiscence and necrosis, including full-thickness loss of dermal tissue extending well beyond the original incision. Also, the positioning screw had lost its fixation (see sixth photo to left). We took the patient back to the operating room, debrided the skin edges, and discovered infection and necrosis of the Achilles tendon. We removed the positioning screw and posterior locking plate and screws due to their involvement. The graft underwent debridement but overall appeared unaffected. Deep bone cultures and bone biopsy also took place.

Clinical Pearls and Pitfalls

Looking back at this case, infection resolution became extremely difficult. Despite 6 weeks of intravenous antibiotics, the initial surgical wound continued to drain a serous type of joint fluid. Cultures remained negative at this point, more than likely due to intravenous care. Thus, it was deemed “a wound problem” where granulation over the bone PMMA spacer did not occur despite NPWT.

Referral to a plastic surgeon and a rotation flap ultimately solved the wound problem. The need for relaxation laterally for flap healing left the unstable ankle in severe valgus, with the foot in a sidecar position. The ultimate problem now was to restore the tibial to floor height and create a space for the lost talar body. Options included a bioengineered talus created by CT comparison to the opposite ankle. However, this did not appear to be a good choice with the prolonged presence of infection. Thus, we chose an allograft, as insurance issues were also a problem. A posterior approach to the ankle offered the best angiosome corridor, and the procedure went extremely well.

Soon after, another postoperative infection occurred, despite the original wound having healed nearly 6 months prior. We feel that likely, a biomembrane (glycocalyx) formed at the bone interface. Eukaryotic bacteria can live in a sessile state for years, and once an operative insult occurs, an explosion of bacteria can extend into the reconstruction site. This led to further complications; however, the patient is well on her way to limb salvage (see last photo to left). Because the talus is a keystone between the leg and foot, timely identification and surgical management are critical in preserving lower extremity function. 

Dr. Visser is the Director of the Foot and Ankle Surgery Residency program at SSM Health DePaul Hospital in St. Louis. He is a Fellow of the American College of Foot and Ankle Surgeons.

Dr. Smith is a first-year resident at SSM Health DePaul Hospital in St. Louis.

1. Looze CA, Capo J, Ryan MK, et al. Evaluation and management of osteochondral lesions of the talus. Cartilage. 2016;8(1):19–30.

2. Hyer CF, DeCarbo WT. (2013). Talar Avascular Necrosis. In: Southerland JT, Boberg JS, Downey MS, Nakra A, Rabjohn LV. McGlamry’s Comprehensive Textbook of Foot and Ankle Surgery (4th ed). Wolters Kluwer;2013:1890-1913.

3. Pearce DH, Mongiardi CN, Fornasier VL, Daniels TR. Avascular necrosis of the talus: A pictorial essay. RadioGraphics. 2005;25(2):399–410.

4. Dale JD, Ha AS, Chew FS. Update on talar fracture patterns: a large level i trauma center study. Am J Roentgenol. 2013;201(5):1087–1092.

5. Buchan C, Pearce D, Lau J, White L. Imaging of postoperative avascular necrosis of the ankle and foot. Sem Musculoskelet Radiol. 2012;16(03):192–204.

6. Dhillon MS, Rana B, Panda I, Patel S, Kumar P. Management options in avascular necrosis of talus. Ind J Orthop. 2018;52(3):284–296.

7. Hawkins LG. Fractures of the neck of the talus. J Bone Joint Surg. 1970;52(5):991–1002.