Ankle osteoarthritis: Toward new understanding and opportunities for prevention and intervention

Abstract

The ankle infrequently develops primary osteoarthritis (OA), especially when compared to the hip and the knee. Ankle OA instead generally develops only after trauma. The consequences of end-stage ankle OA can nonetheless be extremely debilitating, with impairment comparable to that of end-stage kidney disease or congestive heart failure. Disconcertingly, evidence suggests that ankle OA can develop more often than is generally appreciated after even low-energy rotational ankle fractures and chronic instability associated with recurrent ankle sprains, albeit at a slower rate than after more severe trauma. The mechanisms whereby ankle OA develops after trauma are poorly understood, but mechanical factors are implicated. A better understanding of the prevalence and mechanical etiology of post-traumatic ankle OA can lead to better prevention and mitigation. New surgical and conservative interventions, including improved ligamentous repair strategies and custom carbon fiber bracing, hold promise for advancing treatment that may prevent residual ankle instability and the development of ankle OA. Studies are needed to fill in key knowledge gaps here related to etiology so that the interventions can target key factors. New technologies, including weight bearing CT and biplane fluoroscopy, offer fresh opportunities to better understand the relationships between trauma, ankle alignment, residual ankle instability, OA development, and foot/ankle function. This paper begins by reviewing the epidemiology of post-traumatic ankle OA, presents evidence suggesting that new treatment options might be successful at preventing ankle OA, and then highlights recent technical advances in understanding of the origins of ankle OA to identify directions for future research.

1 |. INTRODUCTION

It is well understood that the ankle seldom develops primary osteoarthritis (OA),^1–5 especially compared to the hip and knee,^6,7 but instead most commonly (70–78%) develops post-traumatic OA (PTOA).^8,9 In the case of severe trauma, such as intra-articular tibial pilon fractures, important links have been established between altered joint mechanics and PTOA risk.^10,11 Other data suggest however that more than 60% of ankle OA cases develop after lower-energy ankle injuries, more specifically rotational malleolar fractures and ankle sprains.⁹ Ankle fractures and sprains are common, estimated to account for 179/100,000 patients¹² and 3.29/1000 person–years,¹³ respectively. Given the high incidence of these injuries and the fact that they are often treated as simple and benign, determining which specific types of these injuries most frequently lead to ankle OA is an important target to improve health care and cost utilization, with broad societal implications. The underestimation of the potential consequences of these injuries can lead practitioners and patients to accept suboptimal treatment. This is a missed opportunity, because the consequences of end-stage ankle OA, regardless of the predisposing injury, can be extremely debilitating (Figure 1), with associated impairment comparable to that of end-stage kidney disease or congestive heart failure, certainly equivalent to that of end-stage hip OA.^14–16

FIGURE 1.

Appearance of presenting injury radiographs and those from end-stage ankle osteoarthritis in three different cases, each with a different precedent trauma: (A) pilon fracture, (B) ankle fracture, and (C) ankle instability.

The mechanisms whereby ankle OA develops following trauma are poorly understood, but mechanical factors such as residual malalignment, articular malreduction, and ligamentous instability are all implicated.^10,17,18 In contrast to high-energy injuries such as tibial pilon fractures, the level of articular damage and chondrocyte death at the time of the injury for lower energy ankle injuries is relatively modest.¹⁹ This may lead treating physicians to miss important other elements of the injury that impart PTOA risk, such as malalignment or residual instability of the ankle joint.²⁰ When malalignment is a primary concern, supramalleolar osteotomy may offer an opportunity to restore alignment and improve the intra-articular load distribution, thereby mitigating PTOA risk.²¹ Optimizing the diagnosis, staging and treatment of rotational and low-energy ankle injuries, particularly by minimizing residual deformity and instability, could significantly impact outcomes and decrease the incidence of ankle PTOA. A better understanding of the etiology of ankle OA, including the exact types of injuries and time to joint degeneration, can spur new prevention strategies to mitigate ankle arthritis.

A variety of lower extremity injuries are implicated in the origins of ankle OA, including tibial shaft fractures, pilon fractures, low-energy ankle fractures, talar fractures, and ankle sprains. Previous studies of ankle OA epidemiology have attempted to broadly categorize these injuries, without using specific injury/fracture type classifications.^8,9 This limits the utility of the findings for counseling individual patients. Further unknown is the time between injury and the development of end-stage ankle OA. The relatively low prevalence and the substantial length of time it takes for OA to develop, combined with a lack of radiographic and clinical correlation of symptoms related to ankle OA, makes studying the epidemiology of this condition particularly challenging.²² This knowledge will both provide patients and providers with a better idea of what to expect following these ankle injuries and help in evaluating the success of new treatment approaches aimed at mitigating ankle OA/PTOA risk.

There is broad consensus that joint injuries, by virtue of both acute and chronic mechanical factors, initiate a sequence of biologic events leading to PTOA,¹⁷ and studies have documented a persistent inflammatory microenvironment specifically in the ankle after joint injury.^23–25 Little is known about how different joint injuries influence PTOA risk, though. Clinical trials of new treatments aimed at interrupting this sequence of events face barriers of limited capability to predict PTOA risk and poor early indicators of PTOA development. There remains an urgent need for better functional measures of ankle joint loading, as well as early PTOA imaging markers, because diagnosis, prognosis, and evaluation of therapeutic efficacy all require markers sufficiently sensitive to detect initiation and progression of disease.²⁶

The ability to reliably identify injuries that lead to later end-stage ankle OA can open new opportunities for more effective intervention. This information can support more robust and aggressive staging of these injuries, critical to optimizing treatment protocols that better address residual ligamentous instability and articular malreduction, thereby minimizing the occurrence and/or speed of ankle joint degeneration. New conservative and surgical interventions, including improved ligamentous repair techniques and custom carbon fiber bracing, hold promise for improving treatment outcomes that may prevent residual ankle instability, degeneration, and the development of ankle OA. New technologies, including weight bearing computed tomography (WBCT) and biplane fluoroscopy, offer fresh opportunities to better understand the relationships between ankle trauma, residual ankle instability/malreduction, function, and OA development. However, studies are needed to fill in key knowledge gaps here related to specific etiology so that the interventions can target key factors.

This paper summarizes work presented at a workshop hosted by the Orthopaedic Research Society at their 2023 Annual Meeting in Dallas, Texas. The summary begins by briefly touching upon improved diagnostic and surgical approaches to treating joint injuries for preventing ankle OA, highlights new opportunities presented by custom bracing strategies, introduces new methods for clinically assessing alignment deformities, and ends by identifying important directions for future research.

1.1 |. Diagnostic and treatment considerations to limit PTOA development

There is a status-quo and commonly held understanding among the general population, health care providers, and orthopedic surgeons that low-energy rotational ankle injuries are benign, that ankle sprains rarely lead to residual instability, and that current treatment protocols, surgical techniques, and surgical outcomes for treating rotational ankle fractures are extremely successful. This thinking is unfortunately not supported by available data, which when critically analyzed show that ankle sprains, for instance, frequently lead to residual symptoms. A systematic review found that after an index ankle sprain, 5% to 33% of the patients still experience pain and subjective instability, and after 3 years, up to 34% of the patients reported at least one episode of re-sprain, with many reporting continued instability.²⁷ Low-energy injuries such as ankle sprains and ankle fractures have also been shown to be the main drivers of ankle OA, representing the cause for 45% of a total of 639 patients with end-stage ankle OA.⁸ Optimizing the diagnosis and treatment of these injuries could potentially decrease the frequency of residual post-injury bone malalignment and ligamentous instability, which should in theory minimize arthritic degeneration.

More aggressive investigation of these injuries with weight bearing or conventional CT, as well as MRI, could potentially allow for better recognition of the different injury patterns and severities, as well as identification of more subtle syndesmotic and deltoid instability (Figure 2).²⁸ There is a growing body of literature demonstrating that subtle deltoid and syndesmotic instability is more prevalent than initially thought, with patients that have chronic lateral ankle instability demonstrating arthroscopic signs of deltoid and syndesmotic ligament injury in respectively 40% and 9% of the patients.²⁹ Syndesmotic injuries have also been reported to happen in up to 20% of all ankle sprains,³⁰ and it is well accepted that most of these injuries are not properly assessed and diagnosed by conventional physical exam and conventional radiographic assessment.³¹ Carefully controlled clinical studies are needed to better define when such additional diagnostic imaging may be warranted.

FIGURE 2.

Coronal slice of weight bearing CT showing deformities of the hindfoot along with a 3D CT reconstruction of the bones to demonstrate the complexity of osseous structures. (A, B) Chronic ankle instability following frequent ankle sprains. (C, D) Surgical repair of a severe tibial pilon fracture.

Recent cadaveric biomechanical studies have highlighted a role for repair or reconstruction of the anterior aspect of the deltoid ligament in promoting stability of the ankle in external rotation in the setting of complete deltoid ligament injury.³² Similarly, syndesmotic ligament repair and augmentation of the anteroinferior tibiofibular ligament, as well as posterior malleolar fracture fixation, have also been shown to significantly improve ankle stability in the setting of syndesmotic instability.^33,34 Augmented repair of the anterior talofibular ligament, combined with repair of the calcaneofibular ligament, has also been demonstrated to increase the stability of the ankle joint in anterior translation.³⁵

Improving the treatment of these ankle injuries will require changes to clinical practice. Utilizing a more aggressive diagnostic protocol to investigate ankle fractures and ankle sprains, with a low threshold for CT and MRI assessment, may be required. More detailed and anatomic repair of ankle fractures, along with more frequent and robust repair/reconstruction of residual lateral, deltoid and syndesmotic instabilities, are clinical targets that may decrease the incidence of ankle PTOA attributable to residual malalignment and instability.

1.2 |. A role for custom carbon fiber bracing in the prevention and treatment of ankle PTOA

Surgical fracture reduction has been the mainstay of intra-articular fracture treatment for decades, but even with the best surgical effort, there often remains residual joint incongruity and ligamentous instability that can lead to potentially harmful elevated contact stress. Elevated contact stress-time exposure throughout the stance phase of gait has been shown to be predictive of OA status 2 years following intra-articular fracture and surgical repair.¹⁰ It is for this reason that orthopedic surgeons continue to strive for perfect anatomical reduction and stabilization in the operative treatment of these fractures. This comes at the cost of complication risk and extended time in the operating room.

Surgical intervention is not the only way to influence joint contact stresses, though, and recent advances in bracing provide an attractive complementary treatment solution to manage joint contact stress following intra-articular fracture. Carbon fiber custom dynamic orthoses (CDOs), including the Intrepid Dynamic Exoskeletal Orthosis (IDEO) device design developed at Brooke Army Medical Center, have been used to dramatically improve function and reduce pain for thousands of service members and civilians following traumatic limb injury, including intra-articular fracture.^36–40 CDOs are comprised of a proximal cuff that wraps around the leg just below the knee, a posterior carbon fiber strut that bends to store and return energy, a semi-rigid foot plate that acts as a lever arm to deflect the posterior strut, and in some cases a heel cushion between the footplate and shoe (Figure 3).³⁸ The design characteristics of each CDO component can be varied to influence the forces and motions experienced by the limb,^41–48 which in turn influences activation of muscles that cross the ankle and forces acting on the foot.^42,43,45

FIGURE 3.

Prototypical custom dynamic orthosis device design made from carbon fiber parts.

Limb unloading is a common goal in many CDO fittings, but the specific effects on articular contact stress have not been reported. CDOs have been shown to influence foot loading during gait, with results ranging from limited unloading to greater than 50% reductions in loading reported, depending on the location of interest and CDO provided. Compared to walking without a brace, walking in a CDO typically results in reductions in peak forefoot forces during gait.^49–51

Foot loading and the associated activation of muscles crossing the ankle during daily activities likely influence articular contact stresses in the ankle. As the two largest muscles crossing the ankle joint, the gastrocnemius and soleus muscles have a direct effect on ankle joint reaction forces and the subsequent contact stress in the ankle. At approximately 15% of the gait cycle, the center of plantar foot pressure moves anterior to the ankle joint⁵² and creates an external dorsiflexion moment that is counteracted, primarily, by the gastrocnemius and soleus muscles (Figure 4). When wearing a CDO, the external dorsiflexion moment is counteracted by the CDO reducing the muscle force required. As the center of pressure progresses forward the semi-rigid carbon fiber footplate acts as a lever arm to deform the posterior strut and transfer forces to the proximal cuff away from the foot.⁵¹ These reductions in forefoot loading are likely associated with reductions in plantarflexor muscle activity as the CDO helps to counteract the external dorsiflexion moment during stance. Reductions in muscle activity will reduce joint reaction forces and articular contact stress, which holds promise for reducing the risk of OA development or progression following intra-articular fracture.

FIGURE 4.

The triceps surae generates ankle plantarflexion moments in mid to late stance through the Achilles tendon. Custom dynamic orthoses (CDOs) absorb energy as they are flexed in mid stance and return it in late stance, thereby reducing the moment that must be countered by muscle forces. A reduction in the Achilles tendon force (ATF) thus reduces the joint reaction force (JRF).

While the effects of CDOs on foot loading have been reported previously, one major limitation of these studies has been the inclusion of only one CDO design. The ability to unload or reduce loading at specific regions of the foot will help to reduce pain with weight-bearing, decrease load transfer across the ankle joint, and potentially reduce articular contact stress following intra-articular fracture.^49,51 A better understanding on the effects of different CDO design characteristics on foot loading would help to guide future CDO design and provisioning for individuals at risk for OA development.

1.3 |. Emerging technologies for measuring deformity and instability to inform new interventions

Clinical evaluation of ankle alignment and monitoring for ankle OA presently involve the serial acquisition of weight bearing radiographs scored using the Kellgren and Lawrence (KL) grade.⁵³ However, radiographs are less than ideal, since they capture an obscured 2D projection of a complex 3D structure and pathology. Researchers have studied the utility of low-dose WBCT techniques that mounting evidence suggests will provide more sensitive and responsive measures of joint alignment, stability, and degeneration than plain radiographs, with no substantial increase in cost or time and no meaningful increase in radiation.⁵⁴ WBCT provides greater diagnostic value, not only because of its 3D nature, but also because patients are imaged in a standing physiological weight bearing position that makes it possible to reliably assess joint alignment. WBCT could thus be used to guide earlier preventative therapies that address joint malalignment and residual ligamentous instability with bracing or ligament reconstruction to treat chronic ankle instability following ankle sprains or severe ankle fracture reconstructions.

Statistical Shape Modeling (SSM) has arisen as a computational tool to assess the 3D anatomical shape and deformity of bones throughout the body.^55–62 While this 3D approach is promising, limitations exist in some applications including evaluating individual bones without consideration of articulating joint relationships and lack of clinically interpretable results from model outputs. Researchers have begun to tackle these limitations for the purpose of better characterizing ankle OA by quantifying anatomical shape variation through statistical shape models of the tibiotalar and subtalar joints.^63,64

Statistical shape models constructed from WBCT images have been used to compare the shape of the talus and calcaneus between patients with ankle OA and healthy controls (Figure 5). Notable differences in talar shape in the OA patient group included a flattened talar trochlea, prominent posterior process, and widened anterior tibiotalar articulation. Notable differences in calcaneus shape in the OA group included increased surface area and flatter superior articular curvature of the posterior facet. Further investigation of deformities that present along with end-stage ankle OA can differentiate between the alignment of the tibiotalar and subtalar joints. Varus ankle malalignment is observed in the majority of ankle OA patients, where the subtalar joint is aligned in valgus to compensate for the varus tibiotalar joint, whereas non-compensated ankle OA patients remain in an overall hindfoot varus deformity.^9,65–68 In a multi-bone SSM that captures the posture of the foot and ankle, notable differences included a laterally tilted talus, prominent calcaneal anterior facet, narrower calcaneal middle facet, and broadened calcaneal tuberosity in the compensated group compared to non-compensated group (Figure 6).

FIGURE 5.

Group mean comparison of shape differences in the talus (top 2 rows) and calcaneus (bottom 2 rows) of patients with ankle osteoarthritis and similarly aged-matched healthy controls. Group differences visualize osteoarthritis morphology projected on the control mean shape.

FIGURE 6.

Group mean comparisons measured with multi-bone statistical shape models of the calcaneus and talus together of patients with ankle osteoarthritis with either non-compensated or compensated alignment of the subtalar joint. Group differences visualize the non-compensated group projected onto the compensated group mean shape.

Ankle malalignment increases OA risk, leads to more rapid cartilage loss, and increases pain and functional impairment.⁶⁹ The segmentation and longitudinal registration of bones from WBCT provide a direct means for quantifying joint alignment and for tracking how it changes over time.^70–75 WBCT allows a more complete 3D assessment of the foot and ankle under physiological weightbearing load (Figure 7).^76,77 The use of WBCT-based measures in the evaluation of syndesmotic instability, for example, is well established in the literature.^78,79 To evaluate disturbance of the distal tibiofibular syndesmosis, we have developed methods to automatically quantify syndesmotic incisura volume and distance mapping from WBCT.^80,81

FIGURE 7.

Analyses of joint distance mapping, bone shape, and alignment can all be completed working from weight bearing CT. (A) Varus alignment of a patient with ankle osteoarthritis. (B) Morphology quantification using statistical shape modeling. (C) Joint distance mapping of tibiotalar and subtalar joints in a patient with end-stage ankle osteoarthritis. Notice joint space narrowing on the medial aspect of the ankle joint.

Finally, biplane fluoroscopy is a technique involving two calibrated fluoroscopes capturing dynamic video to assess 3D kinematics of individual bones during activities of daily living. It is emerging as a technique for studying the development and progression of ankle OA. Early work has investigated cartilage contact area⁸² and cartilage strain,⁸³ which are measures that can monitor ankle OA. Others have used biplane systems to study the effect of chronic ankle instability on ankle kinematics and cartilage strain.^84–87 More recently, researchers have used biplane fluoroscopy systems to quantify tibiotalar kinematics after total ankle replacement^88,89 and subtalar kinematics after ankle arthrodesis.⁹⁰ This technique is well suited to study PTOA as it develops over time given the highly precise nature of tracking in vivo bone kinematics that can detect subtle functional differences following traumatic injury to the ankle as an early marker of degenerative joint disease.

2 |. CONCLUSION AND FUTURE RESEARCH DIRECTIONS

Rotational ankle fractures and ankle sprains are generally considered benign injuries. As detailed above, research suggests that more than 50% of end-stage ankle OA happens following these lower-energy injuries. Residual ligamentous instability can potentially explain the development of subsequent arthritis, since differently from pilon fractures for example, the level of articular damage and chondrocyte death at the time of the injury is considerably decreased in rotational, low-energy ankle traumas. This evidence suggests that currently applied treatment protocols for these injuries might be missing hidden instability and associated potential to induce PTOA. A possible explanation is that the missing link may be residual instability of one or more of the three ligamentous complexes of the ankle (lateral, medial, and syndesmotic). Investigating, understanding, and better treating residual instability following these injuries may significantly impact outcomes and decrease the incidence of PTOA.

New intervention strategies, combined with evolving assessment and bracing methods, aim to minimize or prevent progressive tissue damage triggered by joint injury. Despite these exciting new opportunities for intervention, significant barriers remain for these opportunities to yield new clinical approaches. Future work must seriously address these barriers by identifying diagnostic tools for quantifying which low-energy injuries place the ankle at risk for PTOA, exploring the effect of ligament surgical repairs on reducing ankle PTOA, quantifying the change on joint loads and subsequent PTOA from carbon fiber bracing, and solidifying the role of emerging technologies in helping to develop new interventions. A common theme of the following list of suggested future research directions is that they embrace emerging technologies that are providing ways in which to monitor the health of the joint tissues during the subsequent period of healing. Ultimately, understanding the healing period will help improve treatment of patients with joint injuries.

1. Improved quantitative methods for assessing ankle injuries.

2. Registries of patients with ankle joint injuries, clinical studies of PTOA risk factors, and prospective controlled clinical trials to determine the efficacy of treatments of joint injuries.

3. Improved methods for the assessment of ankle joint health in the period following initial treatment are needed to better define the nature of damage observed, and to better understand the implications of the findings in terms of future joint health.

4. Studies of the relationships between residual joint incongruity/instability and OA.

5. A greater appreciation that non- or post-surgical treatment strategies may be able to substantially alter the likelihood of later ankle OA.