Effect of aerobic exercise of different intensity on articular cartilage metabolism in patients with knee osteoarthritis: A randomized controlled trial

Song L, Meng Y · Medicine

The human dose-response study this question needed, and within the range it tested more was better rather than moderate being best. Serum CTX-II and COMP were the primary endpoints and both showed a significant time-by-group interaction (p<0.001). At 12 weeks against their own baselines, CTX-II fell 28.5 percent at high intensity (p<0.001), 18.7 percent at moderate (p<0.01) and 8.2 percent at low (p<0.05), while the control group did not change significantly; COMP fell 22.3 percent at high intensity. WOMAC improved 45.3 percent and pain 52.5 percent in the high-intensity arm, where 73 percent reached the minimal clinically important difference on WOMAC and 80 percent on the pain scale. MRI showed cartilage thickness up 4.2 percent and volume up 3.8 percent at high intensity. The overall dose-response was confirmed at F(3,108)=4.82, p=0.003. No serious adverse events occurred. Transient knee pain affected 2, 3 and 5 participants across the low, moderate and high arms and delayed muscle soreness 2 and 4 in the moderate and high arms, all resolving within 48 to 72 hours with nobody withdrawing for an exercise-related reason. Adherence fell as intensity rose, 92, 89 and 85 percent.

Population

Adults aged 50 to 75 with clinically and radiographically diagnosed knee osteoarthritis, Kellgren-Lawrence grade 2 or 3, pain at least 3 of 10, not already in an organised exercise programme

Intervention

Supervised treadmill walking or stationary cycling three times a week for 12 weeks at 50-60, 60-70 or 70-80 percent of heart rate reserve, 30 minutes per session, heart rate monitored continuously

Comparator

A no-exercise control group; 30 participants per arm

Limitations

Thirty per arm and twelve weeks, so this describes cartilage turnover markers and short-term symptoms rather than disease progression. The primary analysis was per-protocol, with an intention-to-treat sensitivity analysis reported as consistent. Published in a journal of general scope rather than a rheumatology or sports medicine one, and not yet replicated. The ceiling matters as much as the slope: the highest arm ran at 70 to 80 percent of heart rate reserve, so the trial establishes that the down-slope is not inside that range and says nothing about what happens above it.