Rheumatology>Arthritis  –  — Trial shows little extra benefit from intensive program

by John Gever, Contributing Writer, MedPage Today January 23, 2023

In Knee OA, Exercise Is Helpful … But Maybe More in Mind Than Body | MedPage Today

A photo of a male physiotherapist guiding a mature woman through knee exercises in a gym.Une image contenant texte, clipart

Description générée automatiquement

High-dose exercise was only slightly more beneficial for patients with knee osteoarthritis (OA) than a less-intense program in a randomized trial, researchers said.

After 12 weeks of program participation, mean Knee Injury and Osteoarthritis Outcome Score (KOOS) ratings for pain, daily function, and quality of life improved with both regimens, but with no significant differences between them, according to Tom Arild Torstensen, MSc, RPT, of the Karolinska Institute in Huddinge, Sweden, and colleagues.

Self-ratings of ability to perform sports and recreational activities did favor the high-dose regimen, with mean KOOS scores of 47 versus 38 at week 12, but the proportions of patients achieving predefined clinically important improvements from baseline still did not differ significantly, the researchers reported in Annals of Internal Medicine .

Follow-up at 6 and 12 months after the formal exercise programs ended also showed no differences between groups.

These findings were disappointing, Torstensen and colleagues indicated, as they expected to find clearer benefit from the high-dose program, which involved supervised sessions lasting 70-90 minutes three times weekly. The group concluded that the benefits of exercise may be more psychological than physical.

« [T]he use of exercise treatment in chronic pain conditions should be viewed as a form of cognitive therapy, where the goal is to modulate the feeling of pain and thus patients’ thoughts and feelings about it rather than increasing muscle strength and endurance, » the researchers wrote.

While some previous studies had found that relatively intense exercise was more beneficial in knee OA than lower-level activity, Torstensen’s group acknowledged that the overall literature is mixed.

Study Details

In this trial, called SWENOR , Torstensen and colleagues randomized 189 patients 1:1 to the two programs.

The study differed from most previous trials in that the per-minute intensity of exercise was not the focus, but rather the total number of minutes doing the prescribed activities. Patients came from four mid-sized cities in Sweden and Norway.

In both groups, patients came to an exercise facility equipped with various weight-and-pulley machines and performed exercises as directed by trained physiotherapists. The high-dose regimen included 11 different exercises, most involving three cycles of 30 repetitions, plus three periods on elliptical trainers lasting 40 minutes in total.

The low-dose regimen involved four of the same types of exercises, scaled down to two cycles of 10 repetitions, plus 10 minutes on the elliptical machine. This group also performed some non-machine exercises such as squats and step-up/-downs. In all, the low-dose program lasted 20-30 minutes per thrice-weekly session.

Following completion of the 12-week programs, patients were told to resume normal activities; they were neither encouraged to continue with the regimens nor discouraged from doing so if they wished.

Mean patient age was 62, and and just over one half were women. KOOS scores at baseline averaged 57 for pain, 58 for other OA symptoms, 66 for daily function, 31 for sports/recreational ability, and 38 for quality of life.

(Higher scores represent lower disease severity.)

Most patients were taking medications of various kinds, but only about half of these were using the drugs for knee or other musculoskeletal pain.

Just over half of patients in both groups showed clinically important improvements (defined as 10-point increases from baseline) in most KOOS domains.

Besides KOOS scores, outcomes also included visual analogue ratings of pain (including its effects on other aspects of patients’ lives such as sleep) and health-related quality of life. These, too, showed clinically relevant improvements from baseline in both groups but no differences between groups.

Adherence was good in both groups but clearly better for the low-dose regimen (97% vs 78%). Five patients assigned to the high-dose program quit early because it was too time-consuming, compared with one in the low-dose group.

The difference in adherence  may go some way toward explaining why the high-dose regimen wasn’t as effective as the researchers hoped.

Torstensen and colleagues did not address the programs’ cost (they were apparently provided free to participants), nor whether patients were paid to participate. In the U.S., programs like those tested in the trial — thrice-weekly sessions at a fitness center with a personal trainer — would likely cost hundreds of dollars per month.

Thus, the real-world applicability to the general OA population is open to question.

Other limitations cited by the authors included the absence of a no-exercise control group, and losses to follow-up after the formal programs ended.

author['full_name'] John Gever was Managing Editor from 2014 to 2021; he is now a regular contributor.


The trial was funded by the Swedish Rheumatic Fund.

Torstensen and co-authors reported having no relevant financial interests.

Primary Source

Annals of Internal Medicine

Source Reference: opens in a new tab or windowTorstensen T, et al « High- versus low-dose exercise therapy for knee osteoarthritis: a randomized controlled multicenter trial » Ann Intern Med 2023; DOI: 10.7326/M22-2348.