Effect of knee extensor power on knee pain in adults with or at risk for osteoarthritis: the MOST study

Key points

1. 01Knee extensor power is how quickly and forcefully the thigh muscle can extend the knee, and it tends to decline with age earlier than raw strength.
2. 02People in the weakest power quartile had about twice the odds of developing frequent knee pain over 2 years compared with the strongest quartile.
3. 03The link to worsening pain severity was modest and not significant within sex groups, but a significant linear trend appeared in the full cohort.
4. 04The risk of new frequent knee pain tied to low power was more pronounced in males.
5. 05Power is modifiable through training, so the authors suggest power-focused exercise as a possible prevention strategy.

How it was conducted

Design

Prospective cohort study using the Multicenter Osteoarthritis (MOST) Study, with a 2-year follow-up

Participants

Adults aged 45 to 92 with or at risk for knee osteoarthritis; baseline at the 12-year MOST visit plus a new cohort recruited in 2016

Exposure

Baseline knee extensor power measured on a HUMAC NORM dynamometer using isotonic contractions at 40% of 1-repetition maximum, normalized to body mass (watts/kg), grouped into sex-specific quartiles

Outcomes

Worsening WOMAC pain (increase of 2 or more on the 0 to 20 scale at 2 or more of 3 follow-ups) and incident frequent knee pain (yes at 2 or more of 3 follow-ups)

Analysis

Knee-level logistic regression with generalized estimating equations, adjusted for age, race, depressive symptoms, and radiographic OA (plus baseline WOMAC for the pain-worsening model)

What they found

• Worsening pain severity, full cohort: weakest vs strongest power quartile had 1.35 times the odds (95% CI 0.98 to 1.86); significant linear trend across quartiles (p = 0.04).
• Incident frequent knee pain, full cohort: weakest vs strongest quartile had 1.95 times the odds (95% CI 1.22 to 3.05); significant linear trend (p < 0.01).
• Incident frequent knee pain, males: weakest quartile had 2.59 times the odds (95% CI 1.18 to 4.83); significant linear trend (p < 0.01).
• Incident frequent knee pain, females: weakest quartile had 1.58 times the odds (95% CI 0.91 to 2.88); significant linear trend (p = 0.04).
• Worsening pain severity in females (1.36 times the odds, 95% CI 0.79 to 1.90, p = 0.19) and males (1.24 times the odds, 95% CI 0.88 to 2.22, p = 0.18) showed no significant within-sex trend.
• Baseline mean knee extensor power was 3.2 watts/kg (WOMAC sample) and 3.3 watts/kg (frequent knee pain sample); mean age was 63.0 years and mean BMI was about 29 kg/m2.

Limitations

• Observational design shows association, not cause, and residual confounding (for example from body mass) cannot be ruled out.
• The WOMAC threshold for worsening pain was low and may have captured people without true worsening.
• Knee extensor strength was not included due to lack of standardized measures at the baseline visit, so power and strength could not be directly compared in the same cohort.
• Pain is multifactorial, influenced by psychological and social factors not fully captured here.

Why it matters

For patients

Keeping the thigh muscles able to extend the knee quickly and forcefully, not just strongly, may help lower your risk of developing or worsening knee pain.

For clinicians

Consider assessing and training knee extensor power, especially in men, as part of strategies to prevent incident and progressive knee pain in patients with or at risk for knee OA.

For readers

This large cohort suggests muscle power may be a more sensitive predictor of future knee pain than strength alone, though trial evidence on power training for prevention is still needed.