Fred is a 58-year-old patient with pain and swelling in both knees. He has always been active and played collegiate and then club rugby until 15 years ago. Every morning, his knees are stiff and at night they ache, especially after an active day. The x-rays show features of osteoarthritis. He has been taking over-the-counter NSAIDS and has tried ice packs to ease his symptoms with little relief. He asks what more can be done for pain relief.

Osteoarthritis is an inflammatory and degenerative condition of the joints, commonly affecting the hips or knees. Initial therapy involves lifestyle modification (e.g., low-impact exercise, weight loss) and analgesics. Other treatment modalities may be sought when symptoms persist or worsen.

The treatment pendulum for osteoarthritis swings between physical therapy and intraarticular glucocorticoid therapy. Both modalities have been shown to provide symptom relief and improve function. Glucocorticoid injections offer quick relief, involve fewer visits, and rely less on patient adherence, but they can be associated with complications and usually require multiple doses. In contrast, physical therapy may offer more-sustained relief but requires patience and a commitment to the exercises and routine visits.

In a randomized study published in NEJM, Deyle et al. compared physical therapy with glucocorticoid injection for knee osteoarthritis in a cohort of patients in the U.S. Military Health System. The results showed that physical therapy was associated with less pain and greater functional benefit at one year than glucocorticoid injections.

The following NEJM Journal Watch summary explains the study in more detail.

Jonathan S. Coblyn, MD reviewing Deyle GD et al. N Engl J Med 2020 Apr 9

Physical therapy is better than corticosteroid injections.

Physical therapy (PT) and corticosteroid injections are two options that are used widely to treat patients with knee osteoarthritis (OA). To compare their relative efficacy, U.S. researchers randomized 156 patients with symptomatic and radiographic knee OA to receive either PT or steroid injections. Patients who had received either intervention in the preceding 12 months were excluded. During the 1-year study, patients in the injection group received a mean 2.6 injections; patients in the PT group had a mean 12 treatment visits.

The primary outcome was improvement on the WOMAC scale, which assesses both pain and function (score range, 0–240, with higher scores indicating worse OA). In intent-to-treat analysis, improvement at 1 year favored PT over steroid injections: Mean preintervention WOMAC score was 108, and mean scores at 1 year were 56 for the injection group and 37 for the PT group. Nine percent of the PT group received injections, and 18% of the injection group received PT.

Comment: The improvement in both groups was impressive in this study, but the results reinforce the benefit of PT over other modalities. Patients often do not adhere to PT long enough to see benefits, so clinicians can use this evidence to recommend PT over injections for at least 1 year.

Browse more From Pages to Practice »

Krista is a 2019-2020 editorial fellow at the New England Journal of Medicine. She is from Nassau, Bahamas where she is training in general surgery at the University of the West Indies.