Dextrose prolotherapy

Prolotherapy injects hypertonic dextrose — 10–25% sugar solution — into and around the joint. The traditional story, that local irritation triggers a proliferative healing response, is unproven; proposed alternatives include sensorineural modulation (dextrose effects on TRPV1-expressing nociceptors) and plain needle-and-expectation effects. No human trial shows a structural cartilage effect, and claims of cartilage regeneration from prolotherapy clinics have no trial support. What it has is something rarer in this section: blinded separation from saline.

The two blinded trials against saline

Rabago 2013 randomized 90 adults with painful knee OA to intra- plus peri-articular dextrose sessions, saline injections on the same schedule, or at-home exercise. At 52 weeks the WOMAC composite improved 15.3 points with prolotherapy versus 7.6 with saline and 8.2 with exercise — a difference exceeding the minimal clinically important threshold, assessed with blinded, injector-independent measurement.

Sit 2020 tested intra-articular dextrose alone in 76 Hong Kong primary-care patients against saline, blinded. At 52 weeks WOMAC pain improved an additional −10.34 points versus saline (P=.022), and the patient-reported function and quality-of-life scores moved with it. The objectively measured function tests did not separate.

A 2022 meta-analysis (14 RCTs, 978 patients) supports the pair: favorable effects on pain, function, and quality of life versus placebo, comparable to other invasive therapies, with signals of a dose-dependent pain response — while noting the trials are small and heterogeneous.

The limits of the evidence base

The entire randomized base is well under a thousand patients across varying protocols — intra-articular alone versus combined intra- and peri-articular, different concentrations, different schedules. Rabago co-authored both flagship trials, so fully independent replication is thin. Several trials in the meta-analytic pool carry risk-of-bias concerns. And the "placebo" cuts both ways: saline injections may themselves be mildly active (dilution, lavage), which would bias toward the null — but unblinded co-interventions and expectation effects in small trials bias in prolotherapy's favor.

Protocols and cost

Trials used 15–25% dextrose over 3–5 sessions spaced about 4 weeks apart — Rabago injected at weeks 1, 5, and 9 with optional sessions at 13 and 17, while Sit used weeks 0, 4, 8, and 16. Dextrose costs pennies; the expense is clinician time, making this by far the cheapest injection therapy in this section — worth holding in mind when orthobiologics with similar-magnitude evidence cost thousands per course. It is not typically insurance-covered.

Safety

Benign in the trials. Rabago 2013 reported no adverse events. Sit 2020 counted eight serious adverse events across 52 weeks — two on dextrose, six on saline — and judged none of them related to the injections, with one knee replacement in each group. Transient post-injection soreness is the expected course. Standard injection contraindications apply — active infection above all. Diabetics warrant glucose caution given the dextrose load, though systemic absorption is trivial.

What would change the tier

Independent, larger replication outside the founding investigator network; clarity on optimal concentration and intra- versus peri-articular targeting; durability data beyond 52 weeks; and any structural correlate — none exists today.