TO THE EDITOR
Osteoarthritis (OA) is a degenerative joint disease which affects over 240 million individuals worldwide and is characterized by pathological changes in all tissues of the joint[1,2]. Along with the hip and joints in the hand, the knee is one of the most commonly affected joints[3]. The typical knee OA treatment begins first with oral analgesics and injections of corticosteroids and hyaluronic acid[4]. However, many patients ultimately progress to late-stage OA requiring total knee arthroplasty (TKA). Previous work has outlined that the hip, knee, and ankle can contribute to formation of OA in another joint by means of joint alignment or compensatory changes[5]. The aging population and rising prevalence of OA underlines the need for a better understanding of the implications of knee OA and TKA on surrounding joints such as the hip and ankle, as well as how symptoms in surrounding joints affect TKA outcomes[6]. Answering this unmet need, Buterin et al[7] conducted a prospective study of patients in which they evaluated how ankle symptoms prior to primary unilateral TKA affected knee and ankle outcomes after TKA. The objective of this paper is to outline the findings of Buterin et al[7] and provide a commentary on limitations and areas for expansion. We hypothesize that additional studies surrounding the knee and ankle relationship in TKA recovery will improve patient outcomes.
Buterin et al’s study[7] found that patients with lesser ankle symptoms prior to TKA experienced significantly better knee outcomes three months after TKA. Specifically, the Knee injury and Osteoarthritis Outcome Score and knee extension were significantly improved (P = 0.02 and P = 0.01, respectively). In addition, TKA contributed to an improvement in ankle symptoms compared to before arthroplasty. These results highlight the importance of understanding the whole limb when providing OA treatment and draw attention to the need for further exploration of the knee and ankle relationship.
Existing work has explored the relationship between the hip-knee-ankle (HKA) angle and the development of knee OA. This angle is measured between the mechanical axes of the femur and the tibia, and is often used as a metric to understand varus and valgus of the knee joint[8]. Patients with an HKA angle which demonstrates more varus or valgus than normal are more likely to progress to severe OA needing TKA[5,9]. This is because varus or valgus disproportionately loads the medial or lateral tibiofemoral compartment, respectively[9]. Buterin et al[7] pointed to the correction of this alignment through TKA as a reason for improved ankle symptoms after arthroplasty. The discussion of knee and ankle alignment highlights how altered gait mechanics from ankle symptoms may be a potential mechanism for poor TKA recovery. Compensatory loading on the ankle joint in response to either knee OA or TKA could also be an explanation for the findings. However, the authors highlighted that the specific relationship between the knee and the ankle is understudied, therefore more work is needed for a conclusion to be drawn.
The study showed that the knee and ankle impact each other and raised the question of which joint is first responsible for symptomatic degeneration seen in the other joint. Muehleman et al[10] conducted a study which attempted to answer this question by evaluating and scoring the knees and ankles of 545 cadaveric joint donors. They found that knee OA was more severe than ankle OA in 60.8% of left lower extremities and 60.5% of right lower extremities[10]. They claimed this data suggests that while similar and potentially independent mechanisms could be causing the simultaneous knee and ankle findings, it is more than likely that pathology in the knee leads to the formation of pathology in the ankle[10]. Another study by Lee and Jeong[11] showed that ankle OA often occurs secondary to knee OA and can be impacted by alignment changes during TKA. These studies, along with previous HKA research, help explain why Buterin et al[7] see improvement in the ankle after TKA, but also why healthier ankles promote better TKA recovery.
The work done by Buterin et al[7] clearly demonstrated a strong relationship between ankle symptoms and TKA outcomes. The widespread incidence of OA as well as TKA emphasizes the importance of this topic and suggests that this study should be expanded upon. Future work should include a multi-center study to further validate the findings as the authors evaluated only a single center, a point that the authors themselves make in their discussion. In addition, existing literature suggests that completion of activities of daily living in non-western countries typically requires a larger range of motion[12]. Current prosthetics used in joint replacements often are not designed with these activities of daily living requirements in mind[13]. Future work could consider evaluating additional metrics for a diverse set of activities of daily living which could evaluate TKA recovery as it relates to ankle pathology across several global populations. In addition, this could inform future prosthetic design to improve patients’ ability to better perform activities of daily living after arthroplasty.
Buterin et al[7] defined their inclusion criteria as follows: Patients undergoing primarily, unilateral TKA, with advanced primary knee OA with a Kellgren-Lawrence grade of 4. They excluded more complex issues such as secondary post-traumatic OA of the knee as well as ankles with post-traumatic OA. This eliminated unnecessary variables for this initial study but presents limitations and creates an opportunity to explore additional scenarios where the ankle and knee relationship may be relevant. The restriction of Kellgren-Lawrence grade to 4 alone could be expanded to grade 3, as literature supports that patients experience equal benefit from TKA when the knee exhibits grade 3 damage[14]. The authors could also consider evaluating a cohort of patients with post-traumatic OA, as the traumatic injury provides a clear starting point to begin evaluating patients in a future prospective study. Post-traumatic OA is responsible for 12% of all OA cases and 9.8% of all knee OA cases[15], and patients with a previous traumatic knee injury are 4.2 times more likely to develop OA compared to patients without knee injuries[16]. Studies could explore how post-traumatic knee OA might also lead to ankle symptoms, as well as how ankle symptoms may contribute to recovery from TKA on a joint with post-traumatic OA. This may change surgical and rehabilitation protocols to be more broadly applicable.
The article used several established metrics to evaluate the progression of knee and ankle symptoms before and after TKA. The first of these is the American Orthopaedic Foot and Ankle Society Score (AOFAS), of which it was implied that the authors used the ankle-hindfoot version[17]. Buterin et al[7] next used the above-mentioned Knee injury and Osteoarthritis Outcome Score[18]. These are widely used and validated tools for ankle-related and knee-related morbidity, respectively, which incorporate both subjective and objective evaluation of the joint. The authors also employed active range of motion to understand knee mobility and the visual analogue scale as a measure of the subjects’ pain. It appears the authors evaluated these metrics on the knee undergoing TKA and the ankle on the ipsilateral side. One future consideration could be also to score the contralateral ankle and knee to the knee undergoing TKA. Muehleman et al[10] found that ankle joint degeneration is often present on both sides, in particular in response to knee degeneration. Defining contralateral ankle symptoms in addition to ipsilateral symptoms would further the field’s understanding of the impact of limb alignment and gait changes.
In terms of study setup and statistical testing, Buterin et al[7] had a relatively small sample size due to the single center nature of their study. Statistical analysis was appropriate as they used standard testing and included multivariable linear and logistic regression analyses to adjust for potential confounders in patient characteristics such as age and body mass index. The statistical significance threshold of P-value < 0.05 was appropriate given the nature of the study. There is a logical clinical mechanism for the statistically significant trends that were discovered even though the study could be strengthened with additional data.
Limitations of Buterin et al’s study[7] include that it was conducted at a single center in a narrowly-defined patient population, which was previously discussed as an opportunity for further study expansion. Another limitation was that data was only collected preoperatively and three months postoperatively. The majority of TKA patients undergo knee rehabilitation for six months to one year, with the maximum improvement occurring around six months after surgery[19]. Considering this, an expansion of Buterin et al’s study should include a six-month timepoint[7]. Buterin et al[7] only briefly mention the lack of perioperative complications, but a later timepoint would be another opportunity to record any longer term complications such as prosthetic failure[20].
Buterin et al’s study[7] would also benefit from a more explicit discussion of confounding variables which affect the biomechanics of the lower limb, such as subtalar joint alignment and soft tissue imbalances. The subtalar joint is documented to compensate for varus or valgus knee alignment, and this joint’s angle can change in response to TKA[11,21]. While the AOFAS considers ankle alignment, it does so only in broad strokes, and Buterin et al’s argument would be strengthened by analyzing specific alignment angles[7]. The worse TKA recovery in patients with ankle symptoms could be confounded by a worse compensation of the subtalar joint. Soft tissue balance in TKA is critical to joint alignment during knee movement and could lead to pain and joint instability that confound study findings[22]. Patients in the study were given rehabilitation exercises to complete on their own and a referral to outpatient physical therapy, but no actual tracking of rehabilitation adherence or quality was recorded. Varied patient adherence to rehabilitation could confound outcomes.
Despite any study limitations, Buterin et al[7] presented a robust study which adds important information to patient management after TKA. They used strict selection criteria which allow for reliability within the data that they present, even if future work can expand on this. They also performed multivariable linear and logistic regression analyses to adjust for age, body mass index, and American Society of Anesthesiologists score between study groups so they could best draw conclusions from the data. In the future, the data from this study could inform rehabilitation techniques to focus on both the knee and the ankle after TKA to promote a positive outcome for patients.
