Invited Commentary

Considering the latest guideline for dealing with osteoarthritis (OA) published by Osteoarthritis Research Society International (OARSI),¹ there seems to be no doubt about the beneficial effect of exercise for patients with knee OA. However, even with about 50 randomized controlled trials (RCTs) published within the past 10 to 15 years,² no one is able to tell who will benefit from exercise and who will not,³ what kind of exercise is the best for patients with knee OA, what dose (intensity, frequency, and duration) is the optimal to achieve the best effect,^3,4 or which principles for exercise progression to follow and how to secure adherence when exercise is prescribed.

It must be necessary to know what the right recommendations are when analyzing the ability of physical therapists to follow them. The MOVE consensus,⁵ frequently referred to in the article by Holden et al,⁶ formulated 10 “commandments” or recommendations prescribing exercise for patients with knee OA. However, only 4 of the propositions were based on category 1A/B evidence, and 6 propositions were based solely upon category 4 evidence (expert opinion). Even though all 10 propositions seem very reasonable, it may be problematic to say that there are disparities between physical therapists’ current use of exercise and the current evidence when so few of the recommendations are evidence-based.

Holden et al use a case to give the respondents the same background, but an interesting question is: How many correct scenarios and answers could be described based on current evidence? We have recently published an RCT comparing land-based and aquatic exercise between patients with knee OA and a control group.⁷ Our exercise program fulfills the recommended exercise types described in the “Discussion” section of our article (ie, includes aerobic, strengthening, and balance exercises), and the vignette describes very nicely the participants in our study. However, we could not find any effect of either land-based or aquatic exercise, indicating that maybe individual differences should be considered much more carefully when prescribing exercise for patients with knee OA. For some patients, the major problem might be pain, and, for other patients, it might be muscle weakness or perhaps loss of motion. The different problems lead to different symptoms and thus different physical therapy interventions.⁸ In addition, we found that the aquatic exercise group had significant fewer side effects than the group performing land-based exercise, raising the possibility that general exercise may not always be as safe as Roddy et al⁵ suggested. This is further supported by an earlier RCT of exercise for patients with knee OA, where the exercise group experienced a higher amount of joint effusion following exercise compared with a control group.⁹

The fact that patients with knee OA are much less physically active than comparable age-matched individuals without knee OA¹⁰ points to the importance of distinguishing between prescribing exercise in order to increase physical activity level and prescribing exercise because a specific exercise may have therapeutic effects for the patient. A possible reason for the positive effect of exercise for patients with knee OA could be that the exercise program simply encourages increased physical activity. Patients with knee OA are less physically active than age-matched individuals without knee OA,¹⁰ and the self-reported positive effect on function and pain could be caused just as likely by the beneficial effect of being more physically active as it could by a specific effect of the disability and pain related to knee OA. This may especially be true if a generalized exercise program is used and individual differences are not considered. Even though it is not yet evidence-based, different impairments and limitations lead to different symptoms or complaints and thus to different physical therapy interventions.⁸

Physical activity could be divided into 3 different groups: (1) the physical activities we do every day in order to move from one point to another or in order to perform a certain action, (2) exercise or training (ie, the kind of physical activity that is repeated enough times to, for instance, improve a sport performance), and (3) exercise with a therapeutic goal. So far, it could be argued that the evidence is very clear on one point: repeating any kind of physical activity and thereby increasing physical activity level seem beneficial for patients with knee OA. In that respect, the physical therapists in the study by Holden et al were in line with current recommendations.

It is very interesting that only 9% of the physical therapists would use exercise alone, indicating that most of the physical therapists prescribed exercise in conjunction with, for instance, thermotherapy, manual therapy, or electrotherapy. This raises another difficulty when trying to identify the evidence and thus establish treatment guidelines for patients with knee OA. The number of different combinations of modalities is almost endless, and designing studies that compare each combination with a control or another combination is not feasible. The only sensible solution seems to be to formulate well-documented theories and then test the best theories in RCTs.¹¹ In order to do so, we need more research on the mechanism behind the development and worsening of OA of the knee and the possible biomechanical and neuromuscular mechanisms behind the effects of exercise and other modalities.

The ultimate goal must be to deliver the best possible treatment to patients with knee OA. The only way to do so is to deliver evidence-based treatment. Holden and colleagues’ study of the implementation of evidence in practice is an important part of that process. However, as long as we do not know (“no evidence”) who will benefit the most from exercise or from what type of exercise, how much is an appropriate dose, how to describe the progression of exercise, and how to secure adherence, we are not able to tell whether the experts are giving the right recommendations or whether the practitioner is doing the right thing.

• American Physical Therapy Association