Effects of a novel foot–ankle orthosis in the non-operative treatment of unicompartmental knee osteoarthritis

Osteoarthritis (OA) is the most common joint disease, with a prevalence of 6 % [7]. OA prevalence increases with age [1].

In patients with knee OA, the medial joint compartment is more commonly affected than the lateral compartment [7]. During normal gait, except for a brief abduction moment after initial heel contact, the knee joint is subjected to an external adduction moment throughout the stance phase [4, 5, 8–12, 16, 19, 20]. This adduction moment is responsible for the load shift from the lateral to medial compartment. This load shift even occurs in the presence of a valgus knee deformity [3–5, 11, 13, 20–22]. The combination of adduction moment and increased medial compartment loads are thought to be responsible, in part, for the high incidence of medial knee OA.

The amplitude of the external adduction moment applied to the knee depends on the joint’s mechanical alignment and ground reaction forces. In patients with medial knee OA, the medial joint space narrows as a result of cartilage degeneration, which shifts the mechanical alignment of the knee into varus. This shift can result in an even greater external adduction moment [22], unless the patient develops a compensatory gait pattern that involves toeing out. A reduction in proprioception [2, 19] that occurs regularly in patients with OA predisposes the joint to abnormal kinematics. These changes are compounded by an increased external varus moment that shifts more load onto the affected compartment, which may also promote degeneration.

A permanent solution for the correction of lower extremity malalignment is a high tibial osteotomy [17]. However, surgical treatment is always associated with risks such as thrombosis, embolism and infection. Conservative treatment options include the use of lateral wedged insoles [2, 9, 14, 20] and knee unloader braces [5–7, 10–12, 16, 22], which can be adjusted to produce a valgus thrust to unload the medial compartment or a varus thrust to unload the lateral compartment. A systematic review of biomechanical studies concluded that knee unloader braces reduce external moments acting on the knee joint [16].

Randomized controlled trials have shown that knee bracing results in improved knee function compared with no bracing in patients with OA and varus malalignment [10]. However, a side effect of brace treatment is skin irritation caused by the condylar pads or straps, which may lead to treatment non-adherence [22].

A new strategy for correcting varus malalignment is to apply an external valgus force to the knee with an ankle-foot orthosis (AFO) (Fig. 1). This AFO (Agilium Freestep®, Otto Bock, Duderstadt, Germany) consists of a non-flexible insole that is connected to a lever with a pad, which in turn applies a valgus force to the lower leg [21, 22]. Skin irritation should be minimized because of the lack of condylar pads.

Recent biomechanical studies have shown that this new brace concept is effective at reducing the knee abduction moment compared with treatment with insoles alone [6, 21]. However, there are no clinical studies that examine the effects of this new brace.

Here we assessed if this AFO has beneficial effects on the symptoms of patients with unicompartmental knee OA. Our hypothesis was that the use of an AFO improves OA related symptoms as measured with the Western Ontario and McMasters Universities Arthritis Index (WOMAC) in patients with unicompartmental knee OA.

This is the first clinical outcome study of the use of a new AFO for unicompartmental knee OA. The results of this study confirmed our hypothesis. The AFO improved WOMAC total score as well as its three subscores: pain, stiffness and physical function.

Improvements in WOMAC total and subscores were about 50 % after 6 and 12 months, which was roughly the efficacy of studies that evaluated knee unloader braces [3, 5, 7].

The improvement in pain VAS score of about 30 % after AFO use, which was statistically significant at the 6- and 12-month follow-ups, was also comparable to that observed in clinical trials of knee unloader braces [6, 10, 12, 18].

These clinical effects confirm previous biomechanical studies that have shown that an AFO has the potential to reduce the knee adduction moment in patients with medial OA [6, 20]. Fantini et al. [6] found significant decreases in knee adduction moment, knee lever arm and joint alignment in the frontal plane with an AFO. Schmalz et al. [20] found that an AFO that is rigid in the frontal plane significantly reduced the maximal adduction moment of the knee from 0.54 Nm/kg to 0.38 Nm/kg.

The clinical and biomechanical effects of clinical interventions such as unloader braces, which are recommended by the Osteoarthritis Research Society International guidelines for the treatment of knee OA, are well documented [3, 9]. A downside of unloader brace treatment is the high rate of therapy discontinuation. In one prospective randomized trial, the compliance with brace treatment was only 45 % [23]. Causes for the discontinuation of brace treatment include a lack of therapeutic benefit or adverse effects such as abrasions, bruises, sores and blisters at the level of the condylar knee pads [23]. In our study, seven of 23 patients reported a partial discontinuation of AFO use for 14 days or longer during the course of their study participation. In six of these patients the cause for discontinuation was not related to the orthosis (vacation or unrelated hospital stay). However, discomfort or light pressure sores around the ankle were evident in seven patients within the first 4 weeks of AFO use. These problems were managed with additional adjustments to the orthosis by a technician. The second type of adverse events, wear and tear of the shoe in which the AFO was worn, was seen in 14 patients after longer AFO use, typically at the 9- and 12-month follow-up visits.

The optimal daily brace use time has not yet been determined [23]. Van Raaij et al. [23] defined compliance as brace use of 6 h a day. In our study the majority of patients (17/23) wore their orthosis for 4–8 h per day [23].

Our results, in combination with previous biomechanical findings [6], suggest that this new AFO design is a sound alternative to conventional knee unloader braces. However, this study has two important limitations. First, this study is not a prospective randomized trial. We lacked a control or placebo group. Prospective randomized trials are needed to further confirm the benefits of this AFO. However, our study design allowed us to avoid several disadvantages of randomized controlled trials. Lack of generalizability is a major problem for randomized clinical trials (RCTs) [15]. Normally, only a small percentage of the patients can typically be enrolled in an RCT because of strict enrollment criteria [15, 18]. The participation rate of some RCTs is less than 15 % [15]. RCT enrollment can therefore be far from the “real world”, making findings difficult to translate into the general population. In our study the eligibility criteria were not highly selective, and patients who were recruited were representative of the patients in our local community. The second limitation of our study is that the examiner was also the caregiver, and the one who informed the patient about the aims of the study. Our study would have been stronger if the assessor was blinded during outcome measurement.

In conclusion, the results of our study suggest that the AFO evaluated here is effective at significantly reducing the pain and stiffness and improving the physical function of patients with mild to moderate unicompartmental osteoarthritis of the knee. The clinical benefits of this AFO appear to be similar to those demonstrated in clinical trials of knee unloader braces, although our study suggests that overall patient compliance with AFOs may be superior to unloader braces. However, additional prospective randomized studies are needed to further characterize this AFO in the treatment of unicompartmental OA.