Prolotherapy is the injection of a small volume of an irritant solution, most commonly hypertonic dextrose, into painful tendon and ligament insertions or into a joint, with the intention of provoking a local healing response that strengthens the tissue and reduces pain. The name comes from proliferant, reflecting the original idea that the injection stimulates proliferation of new collagen. It has been used for decades in sport and exercise medicine (SEM), and the conditions for which it has been offered or studied include tendinopathy, ligament laxity and osteoarthritis (OA). Dextrose is inexpensive and generally well tolerated in small studies, but efficacy and safety evidence remain limited, and it has never become mainstream because the supporting evidence is weak and inconsistent. This page covers what is injected and why, what the evidence shows, and how to counsel a patient who asks for it.
What is injected, and how?
The classic solution is hypertonic dextrose, prepared at concentrations well above physiological, often in the region of twelve to twenty-five per cent for enthesis and ligament injections and sometimes lower for intra-articular use, usually mixed with local anaesthetic. Concentrations, volumes, target sites and treatment schedules vary substantially between protocols and there is no standardised regimen. Other proliferants such as phenol-glycerine-glucose and sodium morrhuate have been used historically but dextrose dominates modern practice because it is cheap, readily available and well tolerated. One proposed mechanism is that a hyperosmolar solution causes a brief local osmotic and inflammatory stimulus, recruiting inflammatory cells and growth factors and initiating a repair sequence where healing has stalled. A separate, more recently emphasised explanation is an effect on sensory nerve signalling, since dextrose appears to influence pathways involved in neurogenic pain, which might explain analgesia occurring faster than structural repair could. Both explanations remain proposed and unconfirmed.
In one described enthesis protocol, small volumes are injected at several points where tendon or ligament attaches to bone, with the needle contacting bone at each point, a pattern often called peppering. Intra-articular prolotherapy is delivered differently.
Technically, prolotherapy differs from most injections in that it is deliberately delivered at multiple points rather than as a single deposit. The clinician identifies the tender attachments by palpation or on ultrasound, and injects small volumes at each of several closely spaced points along the insertion, usually taking the needle down to contact bone at the enthesis, a pattern often described as peppering. Ultrasound guidance is increasingly used, particularly at deeper or higher-risk sites, and it also allows the operator to confirm the target. Treatment is given as a course rather than a single injection, and protocols commonly use repeated sessions, although the number and interval are not standardised. Patients are warned that a flare of pain lasting a few days after each session can occur; it is not required for benefit and should not be interpreted as proof that the treatment is working. Non-steroidal anti-inflammatory drugs are commonly discouraged around treatment on the theoretical basis that suppressing the inflammatory response would defeat the purpose, although this is reasoning rather than evidence.
What does the evidence show, and how should it be discussed?
The evidence is genuinely mixed and should be presented that way. Several systematic reviews and meta-analyses in knee osteoarthritis have reported statistically significant benefit for pain and function compared with saline injection or exercise alone, and reviews across chronic musculoskeletal pain have concluded that dextrose prolotherapy is supported for some tendinopathies and for knee osteoarthritis. Against that, the underlying trials are small, heterogeneous in concentration, volume, injection sites and number of sessions, and frequently at high risk of bias, so certainty is low. On guidance, the American College of Rheumatology conditionally recommends against prolotherapy for knee and hip osteoarthritis because of limited and heterogeneous evidence. In the UK, prolotherapy is not included within recommended standard management for osteoarthritis and is generally accessed privately, which is not the same as a specific prohibition. The fair summary is that prolotherapy may help some patients, that the effect size and the certainty are both low, and that it is not established treatment.
Corticosteroid aims to suppress a local inflammatory response, whereas a proliferant such as dextrose aims to provoke one. The two have opposite intentions and are not interchangeable.
Counselling therefore matters more than technique. The patient should be told plainly that the evidence is limited and inconsistent, that the American College of Rheumatology conditionally recommends against it for knee and hip osteoarthritis, that a course of repeated sessions is usually needed and usually paid for privately, and that they may complete and pay for that course without benefit. It should not be described as regenerative or as rebuilding ligaments, because that outruns what has been demonstrated. It also should not be offered in place of the core measures that do have evidence, namely progressive loading, weight management where relevant, and appropriate analgesia. Safety is the more reassuring part of the picture, but should not be overstated: published studies report mainly transient local reactions, although sample sizes are insufficient to exclude uncommon serious procedural harms. The usual procedural risks of bleeding, infection and post-injection flare are the main considerations, alongside the standard cautions around anticoagulation and injecting near nerves and vessels. Dextrose itself is not a prohibited substance in sport, but the exact formulation and every co-administered agent must be checked against the current Prohibited List, and injectable glucocorticoids are prohibited in-competition, so a corticosteroid must never be assumed to be interchangeable with a proliferant.
Exam Tips
•Prolotherapy injects an irritant solution, most commonly hypertonic dextrose, into painful tendon and ligament insertions or into a joint to provoke a local healing response.
•The proposed and unconfirmed mechanisms are a brief osmotic and inflammatory stimulus that initiates repair, and a separate effect on sensory nerve signalling that might explain earlier analgesia.
•In enthesis protocols it is delivered at multiple closely spaced points, usually down to bone, and given as a course of repeated sessions; intra-articular prolotherapy is not delivered this way, and schedules are not standardised.
•A post-injection flare lasting a few days can occur but is not required for benefit and does not prove the treatment is working; non-steroidal anti-inflammatory drugs are commonly discouraged on theoretical grounds rather than evidence.
•Evidence is mixed: some meta-analyses in knee osteoarthritis are positive, but trials are small and at high risk of bias, and the American College of Rheumatology conditionally recommends against it for knee and hip osteoarthritis.
•It is inexpensive and generally well tolerated but not established treatment; it is not included within recommended standard UK management for osteoarthritis, is generally accessed privately, and must not displace loading, weight management and appropriate analgesia.