OVERVIEW

An extensive literature search was performed in preparation for this chapter. The majority of the articles assessing operative management of patients with end-stage ankle arthritis are of poor scientific quality. Most of the studies provide Level III and IV evidence. Only a handful of Level I and II studies exist, many of which assess the outcomes of total ankle arthroplasty or compare different methods by which to obtain an ankle fusion. Also, no comparative studies exist for ankle fusion versus ankle arthroplasty.

The two latest techniques described for surgical management of end-stage ankle arthritis are ankle joint distraction and fresh total ankle allografts. The literature published on these new techniques includes Level IV evidence only. Consequently, best treatment recommendations of end-stage ankle arthritis are based on expert opinion and conclusions drawn from comparing the outcomes of different studies.

ANKLE ARTHRODESIS

Ankle arthrodesis for end-stage arthritis was first described by Albert in 1879,³ and for almost 100 years, this was the only option offered to patients with severe ankle pathology. Little information has been published on the topic. Since the 1960s, numerous studies have looked at various aspects of ankle arthrodesis: techniques of fusion, fusion rates, open versus arthroscopic fusion, position of fusion, patient satisfaction, functional outcomes, gait analysis, long-term results with focus on the health of the ipsilateral hindfoot joints, and more recently, comparison with normal control populations. These studies are largely based on Level III and IV evidence.

With regard to technique of arthrodesis, the classic Level IV retrospective study authored by J. Charnley⁴ and published in 1951 was the first to demonstrate that compression across a fusion site optimizes bone healing. Charnley⁴ describe compression with external fixation, and this technique was used for more than two decades. During the late 1970s and 1980s, the use of internal fixation increased in frequency. In 1991, Moeckel and colleagues,⁵ in a Level III study, compared external fixation with internal fixation and demonstrated an increased incidence of nonunion, delayed union, and infection in the external fixation group. More recently, circular external fixators with fine wires under tension are being used with increased frequency. The proponents of this technique suggest that the advantages are earlier weight bearing, ability to correct deformities during the postoperative follow-up phase, respect for soft tissues, more stable fixation, and fewer problems with prominent internal fixation.^6,7 Opponents argue that the procedure is more complex, more expensive, and more labor intensive, with increased complications such as pin-tract infections. To date, no comparative or cost analysis studies have been published. Level III studies have suggested that external fixation is advantageous in the presence of infection, bone loss, talar avascular necrosis, or severe deformity^6–8; however, no comparative studies have been performed.

Studies comparing screw versus plate fixation have found an improved rate of union with screws.^9–14 Screw fixation can be obtained with less soft-tissue stripping. This, combined with better compression at the fusion site, may account for the difference. Holt and coauthors¹⁰ describe a technique using three screws for fixation whereas preserving the medial and lateral malleoli. A fibular osteotomy is performed and, after denuding the cartilage, it is secured to the tibia and talus with compression screws. A second screw is placed from the medial malleolus into the talus, and then a third screw, the so-called home-run screw, is inserted from the posterior malleolus into the neck of the talus. This screw is of primary importance because it stabilizes plantarflexion and dorsiflexion forces (Fig. 70-1). Laboratory studies have shown that two crossed screws create a more rigid construct than two parallel screws.¹⁵ Cadaveric studies have shown that the use of three screws has the advantage of increased compression and better resistance to torque.¹⁶ These variations aside,^{9–11,13,14,16} most surgeons would agree that a minimum of two screws is necessary for adequate stability. Stability can be improved further by adding a fibular strut graft¹⁷; the use of a T-plate has been shown on cadavers to provide the stiffest construct when compared with other types of fixation, but it requires more soft-tissue dissection.^18,19

FIGURE 70-1 Recommended position of fusion is neutral dorsiflexion in the lateral plane and neutral alignment of the ankle in the coronal plane, slight valgus of the calcaneus. Equinus and varus positioning should be avoided.

Arthroscopic ankle arthrodesis was introduced in 1983 by Schneider and popularized by others.^20,21 Since then, numerous articles have been published with the primary focus on surgical technique, time to fusion, duration of admission to hospital, and fusion rates.^20–34 Proponents of arthroscopic ankle arthrodesis advocate shorter operating room time, equivalent fusion rates to open methods, shorter hospital stay, and decreased wound healing problems. Again, these observations are based on Level III and IV evidence. Only two comparative studies,^22,34 both of which are retrospective (Level III), offer only marginal support for the aforementioned advantages of arthroscopic ankle arthrodesis. All agree that the procedure requires familiarity with arthroscopic skills for small joints and is indicated only in patients with minimal deformity at the ankle joint level.

Optimal position of ankle fusion has been studied closely in several Level III retrospective comparative clinical series.^35–38 These studies have demonstrated that patients fused in greater than 10 degrees of equinus have a vaulting gait, increased knee extension and recurvatum, laxity of the medial collateral ligament of the knee, and slower walking speeds. These abnormalities were not observed in patients whose ankles were fused in a neutral position.^36–38 It was also observed that when the hindfoot was in varus, patients reported increased pain and callus formation along the lateral border of the foot.³⁶ Current recommendations are to position the hindfoot in neutral dorsiflexion, slight valgus, external rotation equal to the opposite side, and position the talus directly under or slightly posterior to the midline of the tibia.

Overall patient satisfaction after ankle arthrodesis was good. A number of articles published between 1960 and 2006 demonstrated a high satisfaction rate, good relief of pain, and improved function. Most of these articles were retrospective case series studies with no control group. One Level III intermediate follow-up study compared outcomes of ankle arthrodesis with a normal control population, and it demonstrated that there were substantial differences between the two groups with regard to physical function and pain³⁹; however, many of the patients treated with arthrodesis remained satisfied with their results.

Several Level III long-term studies demonstrated a high incidence of ipsilateral hindfoot arthritis after ankle arthrodesis, particularly of the subtalar joint (Fig. 70-2).^35,40–45 Two articles with more than a 20-year follow-up period demonstrated that more than 60% of the patients continue to be satisfied with their results despite the high prevalence of ipsilateral hindfoot arthritis.^40,46 Only two studies have assessed the preoperative radiographs to determine the presence of preexisting ipsilateral hindfoot arthritis.^46,47 Sheridan and coworkers⁴⁶ demonstrated the presence of preexisting subtalar arthritis in 77.5% of patients with end-stage ankle arthritis and suggested that ankle arthrodesis does not necessarily lead to progressive ipsilateral hindfoot arthritis because it is often present before surgery. Daniels and researchers⁴⁷ identified progression of subtalar arthritis in 4 of 26 (15%) patients in an intermediate follow-up study and progression of calcaneocuboid arthritis from stage 2 to 5 in one patient. Trauma has been identified as the most common causative factor of ankle arthritis.⁴⁸ What is not known is whether the preexisting ipsilateral hindfoot arthritis is the result of the stiffness of the arthritic ankle or is caused by the original trauma, or a combination of both. Regardless of the causative factor, it can be concluded that long-term patients with ankle arthritis or ankle arthrodesis, or both, have a high incidence of ipsilateral hindfoot arthritis with the subtalar joint most commonly affected. Level III evidence exists that the presence of ipsilateral hindfoot arthritis does compromise the functional outcome of the patient after ankle arthrodesis.^45,49 None of the long-term studies demonstrated an increased incidence of knee or metatarsophalangeal arthritis on either the ipsilateral or contralateral extremities.^40,46

FIGURE 70-2 (A) Lateral radiograph of a patient that had an ankle fusion six years ago presenting with a one year history of increasing lateral hindfoot pain. Note the end-stage arthritis of the posterior facet of the subtalar joint. (B) Lateral radiograph of a patient that had undergone an ankle fusion 60 years ago. The talonavicular and the calcaneocuboid joints were fused 30 years ago due to the development of arthritis. Patient currently presents with painful subtalar and navicular-cuneiform arthritis.

Articles that have assessed the effects of an ankle arthrodesis on gait have indicated that the abnormal kinetics and kinematics are most evident through the hindfoot, midfoot, and forefoot, with lesser effects on the knee, hip, and pelvis.^{35–38,47,50,51} All the studies cited earlier are retrospective reviews and considered Level III evidence. Although some variability exists in the results of gait analysis, patients with hindfoot fusions generally have the following gait characteristics: shortened stride length, decreased cadence, slower walking speeds, earlier heel raise during stance phase, increased anterior tilt of the tibia during midstance, increased forefoot plantar pressures with the forefoot ground reactive forces (GRF) shifted posterior during terminal stance, increased hip flexion, earlier knee extension during stance, diminished hindfoot motion, and increased midfoot motion. Studies that have assessed the effects of footwear on gait have indicated that the gait pattern further normalizes with footwear, particularly if there is a slight elevation in the heel and a forefoot rocker^36–38; however, what continues to persist are early heel rise, posterior shift of the GRFs, and early extension of the knee.⁵⁰ Beyaert and colleagues⁵⁰ have suggested that the early heel rise and increased anterior tibial tilt during midstance increases the shear forces to the midtarsal joints and is a possible reason for the increased incidence of ipsilateral hindfoot arthritis after ankle arthrodesis. Weiss and coworkers,⁵² in a prospective, Level II, comparative study, assessed the effects of ankle/hindfoot arthrodesis on the function of the knee and hip. This study was not isolated to arthrodesis of the ankle alone, but it is the only prospective gait study that focused on the kinetic and kinematic outcomes of the ipsilateral hip and knee joints. It demonstrated that the overlying leg joints experience an improvement in joint motion, muscle-generated joint moments, and work during walking after arthrodesis.

ANKLE ARTHROPLASTY

Though the role of ankle arthroplasty remains controversial, interest in the procedure as a viable alternative to arthrodesis has increased substantially in the last decade because of the accumulative effect of several factors: (1) the development of foot and ankle surgery as a subspecialty, (2) improved ankle implants, and (3) the success of arthroplasties in other joints such as the hip, knee, shoulder, and elbow.

As is evident from the previous discussion, ankle arthrodesis is considered by many experts to provide a good functional outcome for their patients. However, the long-term consequences of ankle fusion consist of subtalar arthritis, which produces poor functional outcomes because of the stiff and painful subtalar joint complex. A pantalar fusion to address painful hindfoot arthritis below an ankle fusion produces significant functional limitations.^54,55

It is beyond the scope of this article to assess in detail the types of ankle implants utilized today. In brief, there are fixed bearing (two-component) and mobile bearing (three-component) designs. The two-component designs have the polyethylene that locks into the tibial component, whereas three-component designs have a polyethylene that articulates with the tibial and talar component allowing for motion at both surfaces (Fig. 70-3).

FIGURE 70-3 (A) An anteroposterior (AP) view of a two component agility ankle arthroplasty. (B and C) AP and lateral view of a 3-component Hintegra total ankle arthroplasty.

There has been a gradual trend toward a three-component, mobile-bearing implant that requires minimal bone resection.^56–58 The disadvantages of the two-component design are that more bone resection is required and there is less room for error in positioning of the components.^59,60 However, no studies have compared the outcomes of three-component and two-component designs.