Limb Salvage Procedures for Severe Deformities of the Ankle and Subtalar Joint

1. Introduction

Ankle and subtalar joint fusions include isolated tibiotalar or talocalcaneal joint arthrodesis, tibiotalocalcaneal joint arthrodesis, or astragalectomy with tibiocalcaneal arthrodesis, which is performed when the talus is destroyed or dislocated. Even in healthy individuals, these procedures have a number of complications and risk of nonunions [1,2,3]. The incidence of complications increases in patients with diabetes mellitus, Charcot arthropathy, or osteomyelitis [4,5]. Further increases in complications are associated with impaired vascular supply, smoking, and alcoholism [6,7,8].

Surgical treatment of advanced deformities of the ankle and subtalar joint accompanied by bone destruction, osteomyelitis, or skin defect is still an unsolved medical issue. A number of factors must be considered before surgical intervention, in particular the aetiology and severity of the bone defect, the condition of the soft tissues, comorbidities and patient compliance [9]. The goal of conservative surgery is not to preserve functional mobility of the leg, but to stabilize the vital bone and obtain a weight-bearing plantigrade leg [10]. Reconstruction and stabilization of the ankle and subtalar improves foot stability, reduces skin calluses and defects, and improves quality of life [11]. Therefore, some studies recommend early arthrodesis before the onset of complications [12,13]. On the other hand, in selected cases, amputation of the foot is a legitimate treatment method that can rid the patient of the infected lesion, is less demanding and offers a faster recovery.

2. Materials and Methods

The retrospective study included patients who underwent reconstructive surgery for ankle and subtalar joint deformity accompanied by bone destruction, infection or soft tissue defect at our department between January 2013 and December 2022.

We monitored etiology, age, gender, BMI, comorbidities, risk factors, surgery, reoperations, complications, and arthrodesis union. The union of the arthrodesis was monitored by radiographs, and a CT scan was performed when arthrodesis union was in doubt.

The cohort included 29 patients, 16 women and 13 men, 2 patients were operated bilaterally, so 31 interventions were followed (Table 1).

The etiology of the deformity was Charcot arthropathy in 17 procedures (55%), osteomyelitis in 8 procedures (25%), combination of Charcot arthropathy and osteomyelitis in 3 procedures (10%) and rheumatic foot deformity with skin defect in 3 procedures (10%). The most common comorbidities were diabetes mellitus in 22 patients (71%), hepatopathy in 9 patients (29%), rheumatoid arthritis in 8 patients (26%), and ischemic heart disease in 4 patients (13%). The most common risk factors were active smoking in 6 patients (19%), corticosteroid treatment in 5 patients (16%), immunosuppression in 5 patients (16%) and active alcohol abuse in 4 patients (13%). The mean BMI was 29.6, ranging from 19 to 48.

Before or during surgery, infectious agents were cultured in 14 cases (45%). Staphylococcus aureus was the most common pathogen (6 cases), followed by Enterococcus faecalis and Pseudomonas aeruginosa (2 cases). Postoperatively, the infectious agents were cultured in 10 cases (32%). The most common postoperative pathogens was Staphylococcus aureus (4 cases) and methicillin-resistant Staphylococcus aureus (2 cases).

The level of arthrodesis was chosen based on the location of bone destruction, the location of osteomyelitis, and the condition of the soft tissues. The most common procedure was tibiotalocalcaneal arthrodesis in 12 operated legs (39%), of which 6 procedures used an external fixator for stabilization and 6 procedures used an intramedullary nail, astragalectomy with tibiocalcaneal arthrodesis using an external fixator was performed in 10 operated legs (32%) and tibiotalar arthrodesis was performed in 7 operated legs (23%). In 2 cases, a multistage procedure was preferred. The first multistage procedure was a patient with osteomyelitis of the ankle joint. In the first stage, resection of the articular parts of the tibia and talus was performed with implantation of a cement spacer with antibiotics and stabilization with an external fixator. After 7 months, the cement spacer was replaced with an autograft spongioplasty (Figure 1). The second multistage procedure was performed on a patient with an infected nonunion of the ankle. Initially, resection of the destroyed talus and articular portion of the tibia was performed, and the defect was filled with a cement spacer with antibiotics and bridged using an external fixator. 2 months later the cement spacer was replaced with autologous bone grafts according to the Masquelet technique. This was followed by covering the soft tissue defect with a rotational flap. Five months after the first operation, tibiocalcaneal arthrodesis with a trabecular metal spacer nail was performed (Figure 2).

Statistical Evaluation

Fisher’s exact test in contingency table was used for statistical evaluation. We tested arthrodesis union rate in relation to the etiology and arthrodesis union rate in relation to the level and type of fusion.

3. Results

Arthrodesis fusion occurred in 23 of 31 interventions (74%). In relation to the etiology (Table 2), the arthrodesis healed in 13 of 17 interventions (76%) in patients with Charcot arthropathy, in 7 of 8 interventions (88%) in patients with osteomyelitis or septic arthritis, in 1 of 3 interventions (33%) in patients with a combination of Charcot arthropathy and osteomyelitis, and in 2 of 3 interventions (66%) in patients with rheumatic foot deformity and skin defect. The hypothesis of the independence of fusion healing on the etiology of the disease was not rejected in the statistical evaluation (p=0.377). The statistical evaluation did not find an effect of the etiology on the union rate.

In relation to the level and type of fusion (Table 3), tibiotalocalcaneal arthrodesis stabilized with an external fixator healed in 1 of 6 interventions (17%), tibiotalocalcaneal arthrodesis stabilized with a nail in 5 of 6 interventions (83%), tibiocalcaneal arthrodesis stabilized with an external fixator in 8 of 10 interventions (80%), tibiotalar arthrodesis healed in all 7 interventions and both multi-stage operations also healed. The hypothesis that the union rate of arthrodesis was independent of the level and type of fusion was rejected on statistical evaluation (p=0.0199). The dependence was due to the higher number of nonunions in cases of tibiotalocalcaneal fusions using an external fixator.

The arthrodesis failed to heal in 8 interventions (26%), of which in 3 interventions the nonunion was stable and the patients were able to load the limb with an ankle-foot brace. In 5 patients the limb had to be amputated (16%). In the first case, the limb was amputated below the knee 8 months after the primary operation due to progression of osteomyelitis of the feet. In the second case, the infected nonunion in the subtalar joint did not heal and the limb was amputated below the knee 30 months after the primary operation. In the third case, a patient with Charcot’s arthropathy with soft tissue defect developed osteomyelitis of the calcaneus and 24 months after the primary surgery, the limb was amputated below the knee. In the fourth case of a patient with an overlap syndrome of rheumatoid arthritis and systemic lupus erythematosus with corticotherapy, immunosuppression and ankle arthritis, gangrene of the foot occurred. Despite below knee amputation 9 months after primary surgery, the patient died of generalized sepsis. In the fifth case, the patient was an alcoholic, smoker, with chronic pancreatitis, liver cirrhosis and diabetes, with bilateral Charcot arthropathy. He developed osteomyelitis of the ankle with skin defects on the left side. A below knee amputation performed 2 months after the primary surgery failed to heal, and a reamputation in the thigh was performed 16 months after the primary surgery.

Complicated healing of surgical wounds, phlegmon or abscess occurred in 13 interventions (42%). This was influenced by a high frequency of infection before primary surgery. In 3 cases the defect was successfully healed with Vacuum Assisted Closure (VAC) therapy, but in 10 other cases reoperation was necessary. In one intervention, an occult tibial fracture occurred during tibiotalocalcaneal nail insertion. After attempting conservative treatment, the fracture was stabilized with an intramedullary nail. In total, reoperation was necessary in 11 interventions (35%). The mean time from primary surgery to extraction of the external fixator was 117 days, ranging from 87 to 363 days.

4. Discussion

Recently, there have been only a few studies focusing on ankle and subtalar arthrodesis as an alternative to amputation in high-risk patients. One study presented nonunion of arthrodesis in 8 cases and inflammatory complications in 4 of 18 patients [14]. In another study, arthrodesis nonunion was reported in 38% of patients with diabetes and 27% without diabetes [15]. The literature also reports a high rate of reoperations in cases where the patient met at least two of the risk criteria including immunosuppression, diabetes and obesity [3,16]. In our cohort, we observed infectious complications in 42% and arthrodesis nonunion in 26% of interventions. In 5 patients, the limb had to be amputated. Since these are interventions with a high risk of complications, it is necessary to consider a number of factors such as etiology, infectious focus and comorbidity in preoperative planning [17]. The patient must be able to follow the recommended regimen, keep the surgical wound and external fixator sterile. Compliance can be very demanding for patients with neuropathy. Although amputation offers the prospect of a quick solution, it greatly affects the patient’s self-sufficiency. The ability to maintain balance is already limited due to polyneuropathy and in some cases amputation leads to the loss of the ability to walk independently. Therefore, the patient should be informed that the surgical procedure is fraught with numerous complications, including the risk of limb amputation [11].

Charcot arthropathy of the foot leads to progressive instability, which leads to foot deformity, skeletal protrusion, soft tissue defects, and osteomyelitis [18]. The goal of treatment is to reduce inflammation and maintain the foot in a physiological plantigrade position. In the initial inflammatory phase, the patient is monitored, including repeated X-ray examinations, and the affected foot is protected from weight-bearing. Surgical treatment is carried out after the end of the inflammatory phase in cases where the foot is not in a plantigrade position.

If the etiology of deformity is osteomyelitis, the goal is to control the infection and achieve a plantigrade position of the leg [19]. Initially, broad-spectrum intravenous antibiotics are administered, the abscess is drained, and the necrotic bone is removed. After isolation of the infectious agent, targeted antibiotics are administered. Definitive treatment is performed when the infection is stabilized.

In our cohort, tibiotalar arthrodesis was indicated for isolated tibiotalar joint damage. Tibiotalocalcaneal arthrodesis was chosen when both the tibiotalar and talocalcaneal joints were affected. In severe osteoporosis, tibiotalocalcaneal arthrodesis with an intramedullary nail was also selected in cases of isolated tibiotalar joint destruction, thus favouring better mechanical stability over preserving mobility of the unaffected talocalcaneal joint. Astragalectomy with tibiocalcaneal arthrodesis was chosen in cases of talus destruction or dislocation [20,21,22]. Astragalectomy can correct severe leg deformities, but resection of the talus leads to shortening of the limb [23]. In one relatively healthy patient in our cohort, the talus was replaced with a nail with a trabecular metal spacer and and the limb length was preserved.

In 21 interventions, an external fixator was used to stabilize the arthrodesis. The advantage is that an external fixator inserted into the vital bone can achieve reduction, mechanical stability, and compression at the level of the infected or altered bone [24,25,26]. Skin defects can be dressed more easily and VAC therapy can be applied [27,28]. The disadvantage is the need for local care of the external fixator and the frequent occurrence of pin track infections [18,29]. However, in the group of tibiotalocalcaneal arthrodeses stabilized with an external fixator, we observed the lowest fusion rate, namely in 1 out of 6 procedures. Focusing on this group, the indications for surgery were Charcot arthropathy, septic arthritis, a combination of both and soft tissue coverage was defective. The combination of soft tissue coverage defect with bone destruction may represent a borderline indication for reconstructive surgery. In our cohort, the mean time to removal of the external fixator was 117 days (16.7 weeks), and after removal, the ankle was further fixed with a custom-made plastic brace for 6 to 12 weeks. In other studies, the time to removal of the external fixator ranged from 18 to 27.7 weeks [2,14,30], but there was no further fixation with a brace.

5. Conclusions

Reconstructive surgery for advanced deformities of the ankle and subtalar joint due to Charcot arthropathy, osteomyelitis, or rheumatoid arthritis is associated with many complications. The goal of treatment is to eliminate the infection and to achieve a limb capable of bearing weight in the plantigrade position. Although amputation offers the prospect of a quick solution, in neuropathic patients it can lead to the loss of the ability to walk independently. In our high-risk patient cohort, we preferred reconstructive surgery over amputation and in 26 of 31 interventions, the limb was preserved and weight-bearing capacity was restored.