Prolotherapy for Sacroiliac Joint and Ligament Pain

Prolotherapy involves injecting an irritant solution — most commonly hypertonic dextrose — at ligament and tendon insertions, with the aim of provoking a controlled local healing response. The mechanism is deliberate inflammation followed by repair, which is the opposite of what a corticosteroid does.

That mechanism is not theoretical. A dose-escalating placebo-controlled toxicity study in 48 miniature swine, using a phenol-glycerine-dextrose formulation injected into lumbosacral ligaments, found on histopathology at 24 hours exactly what the theory predicts: haemorrhage, inflammation, necrosis and vascular change at the injection sites. At 14 days the same tissue showed repair under way, with fibrosis and skeletal muscle regeneration. Liver and muscle enzymes rose in a dose-dependent way at 24 hours and had returned to normal by 14 days.

That study used a solution containing phenol, which is not what is used here — plain hypertonic dextrose is. It is cited because it is the clearest available demonstration that the mechanism is real, and because a treatment that works by causing injury deserves to be described that way rather than as something gentle.

The 2025 multispecialty consensus guidelines place dextrose prolotherapy and PRP in the same category: weak evidence supporting at least three months of pain relief in sacroiliac joint complex pain. That is the ceiling of what can be claimed.

Cusi and colleagues in Sydney published the protocol that most ligamentous prolotherapy at this joint descends from: three injections of hypertonic dextrose into the dorsal interosseous ligament of the affected sacroiliac joint, under CT control, six weeks apart, in patients with deficient load transfer at the pelvis.

Twenty-five patients attended at least one follow-up. Function improved significantly on the Quebec Back Pain Disability Scale and the Roland-Morris questionnaires at 3, 12 and 24 months, and clinical scores improved significantly from baseline at all three points. Positive clinical outcomes were recorded in 76 per cent of those attending at 3 months, 76 per cent at 12 months, and 32 per cent at 24 months.

Two features of this study deserve to be read carefully. It was a prospective descriptive series in the authors' own private practice with no control group, so regression to the mean and the natural history of the condition cannot be separated from the treatment effect. And the number of patients returning fell substantially at 12 and 24 months, which makes the later percentages less reliable than the earlier ones.

The design feature worth keeping regardless of the outcome data is the prerequisite: patients completed a three-month stability programme before any injection. That sequencing is the standard applied here.

If the Kim trial is the optimistic end of this literature, Hoffman and Agnish's retrospective cohort is the sobering one, and it should be read alongside it.

One hundred and three patients in a Department of Veterans Affairs outpatient clinic, diagnosed with sacroiliac joint instability, received a series of three injections of 15 per cent dextrose in lidocaine at roughly monthly intervals. They were a difficult group: median two years of low back pain, mean pre-treatment Oswestry Disability Index 54 out of 100.

At a median of 117 days after treatment, 24 of 103 (23 per cent) had achieved a minimum clinically important improvement.

One in four, in a chronic, disabled, real-world population, is a defensible reason to offer a treatment — and it is a very long way from the impression created by quoting the Kim trial alone.

Broader systematic reviews of dextrose prolotherapy across musculoskeletal conditions report supportive findings for spinal and pelvic pain due to ligament dysfunction. The most frequently cited of these was authored from a clinic whose practice is prolotherapy, and it is fair to weigh that when reading its conclusions. It is cited below so that you can see it and judge it, not because it carries the argument.

The narrative reviews of prolotherapy technique are similar: useful for describing how the treatment is delivered, not designed to establish whether it works.