Low-impact cardio

Cartilage needs cyclic load; painful knees often can't take much of it at once. Low-impact cardio resolves the tension: it delivers the fluid-pumping compression-and-release cycle without the impact.

That activities differ in what they cost the joint is measurable in the cartilage itself. Imaging 23 healthy knees immediately after thirty minutes of walking, running and stair activity, T1 rho and T2 values fell after every one — consistent with water being pressed out under load — and fell further after running than walking, and further after stairs than running. The joint keeps a running tab, and the size of the entry depends on the activity.

What that costs in newtons has been measured directly only in replaced knees. Eight people with instrumented implants recorded peak contact forces of 3,372 to 4,218 N across everyday activities, rising to 5,165 N in slow jogging — and the largest twisting moment about the implant came not from jogging but from walking. Those are implanted joints rather than native ones, so they set the scale rather than the number.

What the trials show

Cycling. A network meta-analysis of 39 trials and 2,646 patients then compared five exercise types against each other. Stationary cycling ranked first for WOMAC pain and first for the six-minute walk. Read the ranking for what it is: a probability of being best, not a measured gap between the options, from a network in which most trials could not blind anybody. It supports cycling as a reasonable default, not as a demonstrated winner.

Aquatic exercise. The Cochrane review (13 RCTs, n=1,190, knee/hip OA) found moderate-quality evidence of small short-term benefits — pain SMD −0.31 and disability SMD −0.32, each worth about 5 points out of 100 (95% CI 3 to 8), over programs averaging 12 weeks. No serious adverse event tied to the water was reported in any trial. Real but modest; the main value is enabling movement when land exercise hurts. Note what is not there: not one of the 13 trials imaged a joint, so this review is silent on structure.

Head-to-head. A 48-person RCT compared 12 weeks of swimming against cycling (45 minutes, three times weekly, 60–70% heart-rate reserve). Both groups improved on WOMAC pain, stiffness and functional limitation, gained 15–30% in isokinetic knee strength, and walked 6–8% further in six minutes; for the swimmers the WOMAC falls were around 40%, 30% and 25%. Nothing separated the two modes. Pick the one you'll keep doing — with the caveat that the trial had no non-exercise arm, so those improvements are measured against each group's own starting point rather than against not training. Meta-regression across knee-OA trials backs this up: aerobic exercise matches resistance training for pain relief.

Elliptical. No dedicated RCTs in OA exist. Its inclusion here is an extrapolation from its low-impact profile, and we're saying so plainly.

Dosing and setup

Trials used roughly 30–45 minutes, three times per week, for 8–12+ weeks at moderate intensity. On the bike, setup determines joint load: keep resistance modest and cadence higher to minimize knee force, and raise the saddle within comfort. High-power cycling pushes knee load back toward walking levels — so after cartilage repair, start low-resistance.

Limitations

Aquatic effects are small and short-term with little long-term data, and the cycling evidence is a ranking rather than a measured advantage. The swimming-versus-cycling RCT had no non-exercise control, so the absolute effect is uncertain. The direct joint-load measurements come from eight people with instrumented knee replacements, not native or repaired knees, and none of them was cycling. The cartilage-imaging comparison is 23 healthy young knees measured immediately after a single session, so it describes the acute response to load rather than what repeating it does. And no cardio mode has been shown to alter cartilage structure — the demonstrated benefits are symptoms, function, and fitness.

Safety

Adverse-event rates were low across trials. Two practical cautions, both biomechanical inference rather than trial-tested findings: keep early post-repair cycling at low resistance, and note that deep knee flexion at the top of the pedal stroke can irritate patellofemoral lesions.

Open questions

Elliptical trials in OA; whether long-duration low-impact cardio slows structural progression; and the optimal cycling dose after MACI or microfracture.