Fast reps with lighter weights work best for arthritic knees

Exercise is a mainstay of managing knee osteoarthritis, yet the advice usually stops at do some resistance training. That leaves the real questions unanswered, because resistance training comes in many flavours: slow and heavy, light and fast, and everything between, with doses ranging from a handful of repetitions a week to hundreds. Which combination actually eases pain, loosens stiff joints and restores everyday function has never been clearly pinned down. This review set out to rank the training types against each other and to identify the intensity, programme length and weekly repetition count that deliver the most symptom relief.

The researchers searched five major databases from inception to April 2025 for randomised controlled trials of exercise in people with knee osteoarthritis. They then ran a Bayesian network meta-analysis, a technique that ranks multiple treatments against one another even when they were never tested head to head in the same trial. The final analysis covered 46 randomised trials with 3463 participants. The outcomes were pain, stiffness and physical function, and the team also modelled dose-response relationships to estimate the best training intensity, duration and number of weekly repetitions for each symptom.

High-speed resistance training came out as the most effective type across the board, with large improvements in pain, stiffness and function compared with the alternatives. The dose-response analysis added detail: for pain and function, moderate-intensity training at 43 to 47% of one-rep max, sustained for 35 to 37 weeks with 610 to 640 repetitions per week, appeared optimal. Stiffness followed a different pattern, responding best to a shorter, denser prescription of around 12 weeks at 1200 weekly repetitions. In other words, no single dose suited every symptom, and the authors argue programmes should be tailored to what a person most wants to fix.

That exercise helps knee osteoarthritis is well established; what this analysis adds is a ranking of training styles and concrete dose targets, which most guidance lacks. The finding that lifting at speed with moderate loads beats other approaches complicates the common assumption that training around arthritic joints should be slow and gentle. Network meta-analyses lean partly on indirect comparisons, though, and the included trials differed in their populations and protocols, so the tidy dose windows are best treated as estimates. Head-to-head trials pitting high-speed training directly against traditional lifting would firm the ranking up.

For anyone with creaky or arthritic knees, this is a reason to think the style of leg training matters, not just whether it happens. Moving moderate loads quickly, with plenty of total repetitions, was the pattern most associated with less pain and better function here. It is also worth noticing that the optimal pain and function doses ran over many months, so a few weeks of training may undersell what a full programme can do. None of this replaces individual guidance, but it sharpens the questions worth asking about a knee-friendly programme.