Key Finding
Acupuncture provided clinically meaningful pain relief and functional improvement in knee osteoarthritis patients for 3 to 6 months after the completion of treatment, outperforming sham acupuncture, usual care, and diclofenac across multiple trials.
Knee osteoarthritis (KOA) is a common condition that causes chronic joint pain, stiffness, and difficulty moving — and many standard treatments only offer short-term relief. Researchers wanted to know whether acupuncture could provide lasting benefits even after the treatment sessions are finished.
To find out, scientists reviewed 10 high-quality clinical trials involving 3,221 people with knee osteoarthritis. They looked at pain levels, joint function, stiffness, and overall quality of life at multiple time points after acupuncture treatment ended.
The results were encouraging. Compared to sham (fake) acupuncture, real acupuncture showed meaningful improvements in pain and function lasting up to 4.5 months after treatment ended. When compared to usual care — such as standard medications or physical therapy alone — acupuncture's benefits lasted up to 5 months. Perhaps most impressively, acupuncture outperformed the common anti-inflammatory drug diclofenac, with benefits lasting up to 6 months after the final session.
For patients who received no treatment at all, one large study found acupuncture's benefits held for 3 months post-treatment, though results across studies were mixed beyond that point. One important nuance: when acupuncture was added to exercise-based physical therapy, it did not show additional advantages over exercise therapy alone in the longer term.
Acupuncture was also found to be safe, with no serious side effects reported across the studies reviewed.
What does this mean for you? If you're living with knee osteoarthritis and looking for a treatment that may offer relief well beyond your last appointment, acupuncture is a promising, evidence-backed option worth discussing with your healthcare provider. To get the best results, seek care from a licensed, board-certified acupuncturist with experience treating musculoskeletal conditions.
This systematic review and meta-analysis evaluated the durability of acupuncture's therapeutic effects in knee osteoarthritis (KOA) following treatment completion, pooling data from 10 RCTs with 3,221 participants. Primary outcomes — WOMAC pain and function subscales — were assessed at all available post-treatment time points. Versus sham acupuncture, clinically meaningful improvements in pain and function persisted for approximately 4.5 months post-treatment. Versus usual care, durable benefits were observed up to 5 months; versus diclofenac, up to 6 months. Evidence for acupuncture versus no treatment was mixed, with one large trial showing 3-month durability and another showing no significant effect at 9 months. Acupuncture as an adjunct to exercise-based physical therapy demonstrated no superiority over EPT alone at up to 11.25 months post-treatment. The safety profile was favorable across included trials. Clinical takeaway: standalone acupuncture offers durable, clinically important KOA symptom relief for 3–6 months post-treatment, supporting its use as a primary intervention rather than adjunctive therapy.
Browse our directory of verified licensed practitioners near you.
Find a practitioner →📌 A six-week course of electroacupuncture produced a clinically meaningful reduction in WOMAC global scores compared to sham acupuncture (MD: 40.56; Cohen's d=1.21), with exploratory MRI data suggesting possible anti-inflammatory and cartilage-protective structural effects.
📌 Twelve sessions of electroacupuncture over four weeks reduced hand osteoarthritis pain significantly more than topical diclofenac sodium gel, with a between-group VAS difference of 14 points and benefits persisting through a 12-week follow-up.
📌 After six weeks of twice-weekly electroacupuncture, a statistically significant proportion of ankylosing spondylitis patients achieved at least a 30% reduction in pain severity scores despite inadequate relief from conventional analgesia (p = 0.002).