Is PT good for knee OA?

Contrary to routine recommendations, physical therapy (PT) does not appear to benefit individuals with knee pain due to degenerative meniscal tear, findings from the TeMPO* trial suggest.

“We found that the combination of home exercise (HE) and PT sessions did not result in greater pain reduction over 3 months than HE alone. The addition of motivational text messages (TM) to HE also did not improve pain outcomes over HE alone,” the investigators said.

The study included 879 participants (mean age 59.2 years) with knee pain, osteoarthritis (OA), and meniscal tear. They were randomized 1:1:1:1 to a 3-month HE programme (group 1), HE + TM (group 2), HE + TM + sham PT (group 3), and HE + TM + standard PT (group 4). The mean KOOS pain subscore was 46.1. Two-thirds of the participants had a Kellgren–Lawrence grade of 0–2. [N Engl J Med 2025;393:1694-703]

The primary outcome was the change in Knee Injury and Osteoarthritis Outcome Score (KOOS) pain subscore from baseline to 3 months. The differences between groups 1 and 4 and between groups 2 and 4 were similar (2.5 points). The difference between groups 1 and 2 was -0.1 points; between groups 3 and 4, it was -0.7 points.

All results failed to reach the minimal clinically important change, which was a decrease in KOOS pain subscore by 8 points, the researchers noted.

A total of 409 participants did not have treatment failure at 3 months and had 12-month data available. Of these, ~80 percent met the criteria for treatment durability, which was defined as “maintaining at least an 8-point improvement in KOOS pain subscore at the 12-month follow-up while not receiving an intra-articular injection or knee surgery”. There were also no meaningful differences across groups in KOOS pain and Activities of Daily Living subscores, and strength and performance tests in this subgroup.

At 6 and 12 months, the adjusted differences in the change in KOOS pain subscore from baseline between groups 1 and 4 were 4.1 and 2.5 points, respectively.

Of note, the changes in KOOS pain subscores in groups 3 and 4 were essentially identical across all timepoints. “These suggest that contextual effects are likely to explain the small apparent differences in pain between groups 1 and 4 over 12 months,” they noted.

One participant in group 4 died. Overall, 33 participants were hospitalized and 80 had arthroscopic partial meniscectomy on the index knee over 12 months. There were few (<2 percent) cardiovascular, pulmonary, infectious and neurologic adverse events.

Interventions

Treatment guidelines do suggest PT or supervised exercise as first-line treatment for knee pain and degenerative meniscal tear. However, the study shows otherwise, as the addition of PT or TM to encourage adherence to HE was not superior to HE alone for pain reduction, the researchers said.

HE involved muscle stretching (quadriceps, hamstring, gastrocnemius) and strengthening (gluteus maximus and medius, quadriceps, hamstring). Instructional pamphlet and video, ankle weights, and guidelines for progression of weights were provided.

In the TM groups, three TM with theory-based statements were sent weekly and pamphlets were mailed twice monthly for 3 months to encourage adherence.

Sham PT included knee symptom assessment, knee ultrasound, inert lotion application along the midthigh and distal tibia, and in-clinic sham manual therapy (minimal force to non-articular areas; no joint mobilization). “Although sham PT is not a true ‘placebo’, our intent was to craft an intervention that controlled for interpersonal attention without having plausible biomechanical effects,” the researchers explained.

Each standard PT session followed an unsupervised warmup on a stationary bicycle and included manual therapy (soft tissue and joint mobilization, stretching of tissues around knee) and therapist-directed strengthening and functional exercises (gluteus maximus and medius, quadriceps, hamstring).