Home exercise matches clinic physical therapy for pain

In degenerative meniscal tear with osteoarthritis, adding clinic physical therapy or text reminders did not reduce pain beyond a structured home-exercise program at 3 months. Four-arm multicenter randomized, sham-controlled trial; Level 1b (2011 OCEBM).

Background

Physical therapy is widely recommended for knee pain with degenerative meniscal tear, but its added benefit over home exercise and contextual effects has been uncertain.

Patients

  • Adults 45–85 years with knee pain, MRI-confirmed meniscal tear, and radiographic OA (Kellgren–Lawrence 0–3).
  • N=879; mean age 59; baseline KOOS pain 46 (0–100, higher=worse).

Intervention

  • Home exercise + text messages + standard PT (14 supervised sessions: manual therapy + strengthening/functional/stretching).
  • Home exercise + text messages + sham PT (attention control: sham ultrasound/manual therapy).
  • Home exercise + text messages (adherence prompts).

Control

  • Home exercise alone (100 minutes/week with instructional materials and ankle weights).

Outcome

  • Primary: 3‑month change in KOOS pain.
  • Secondary: KOOS ADL, strength/performance tests, treatment failure (no ≥8‑point pain improvement or injection/surgery), 12‑month durability, EQ‑5D, adverse events.

Follow up period

Primary 3 months; secondary 6 and 12 months.

Results

Outcome Comparison/Metric Result CI NNT
Primary
KOOS pain change (3 mo) Home exercise vs +Text −0.1 points 98.3% CI −3.8 to 3.7
KOOS pain change (3 mo) Home exercise vs +Text+Standard PT 2.5 points 98.3% CI −1.3 to 6.2
KOOS pain change (3 mo) +Text vs +Text+Standard PT 2.5 points 98.3% CI −1.4 to 6.5
Secondary
Treatment failure (3 mo) Rates (%) Home 36.2; +Text 31.6; +Text+Sham 29.5; +Text+Standard 34.8
Treatment failure vs Home Absolute risk reduction +Text 4.6%; +Text+Sham 6.7%; +Text+Standard 1.4% 22; 15; 71
Arthroscopic partial meniscectomy (12 mo) Rates (%) 9.6; 9.5; 8.2; 9.1 (similar across groups)

Limitations

  • Predominantly White cohort; generalizability limited.
  • Participant blinding imperfect; contextual effects likely.
  • COVID-19 disruptions; 30‑minute PT sessions; some secondary analyses not multiplicity-adjusted.

Funding

U.S. NIH (NIAMS) and Australian NHMRC; minimal industry bias.

Citation

Katz JN, Collins JE, et al. A Randomized Trial of Physical Therapy for Meniscal Tear and Knee Pain. N Engl J Med. 2025;393:1694-703. DOI: 10.1056/NEJMoa2503385. TeMPO; NCT03059004.