What Is It?
Prolotherapy is a commonly offered alternative treatment for joint, connective tissue, and back pain. The practice consists of injecting substances into painful or dysfunctional joints and connective tissue with the intent of relieving discomfort and restoring function. There is a wide range of substances that are used with a much variation among practitioners of prolotherapy. Examples of these substances include: dextrose or other sugars, Vitamin B12, local anesthetic agents, a wide variety of herbal products, homeopathic remedies, zinc sulfate, the patient’s own blood, and hundreds of others. Often, cocktails of multiple substances are used based on the individual preferences of the practitioner.
The theoretical rationale for the practice is that the substances injected into damaged joints or connective tissue will cause inflammation and chemical or cellular activity that will lead to repair of the affected tissues. (1,2) The logic of this theory is questionable. For one thing, inflammation is a normal physiologic response to tissue injury, and is already present in damaged tissues and joints. However, the evidence is clear that excessive or prolonged inflammation interferes with healing. Arthritis, a disorder prolotherapy is supposed to treat, is by definition chronic inflammation of joints, and this inflammation is known to cause much of the pain and tissue damage associated with arthritis. In general, excessive or chronic inflammation does more harm than good, and many conventional and alternative medical therapies are intended to reduce the deleterious effects of inflammation, so deliberately causing it to induce healing seems a dubious approach.
There are theoretical arguments involving the stimulation of cells involved in healing, such as fibroblasts, or the release of growth factors and other substances that are associated with tissue repair, which might plausibly explain how prolotherapy could be beneficial. However, the in vitro research to date suggests the effects of prolotherapy agents on tissues are broad and non-specific, and such effects are as likely to be harmful as beneficial. The underlying theory is questionable but not impossible. However, without a clearly demonstrated physiological mechanism and consistent animal model and clinical research evidence, it is not enough to justify the use of this therapy or the claims often made for it.
Does It Work?
There have been a number of studies and reviews investigating the use of prolotherapy in humans, and more are ongoing. Overall, the evidence has been mixed and of generally low quality. A Cochrane Review of prolotherapy for lower back pain examined five studies involving 366 people.(3) The review concluded,
There is conflicting evidence regarding the efficacy of prolotherapy injections for patients with chronic low-back pain. When used alone, prolotherapy is not an effective treatment for chronic low-back pain. When combined with spinal manipulation, exercise, and other co-interventions, prolotherapy may improve chronic low-back pain and disability. Conclusions are confounded by clinical heterogeneity amongst studies and by the presence of co-interventions.
The insurance companies Aetna and Cigna and the Veterans Administration have published extensive reviews of the scientific literature on prolotherapy.(4,5,6) The reviews conclude:
Aetna considers prolotherapy (also known as proliferant therapy or proliferation therapy) experimental and investigational for any indications because there is inadequate evidence of its effectiveness.
Medical studies in the literature evaluating this technology present few randomized, double-blind clinical trials that had adequate sample size and controls and that used objective outcome measures. Additionally, studies have not successfully supported the use of prolotherapy as an effective treatment for joint or ligament instability. Furthermore, several systematic reviews, a Cochrane review, and technology assessments have reported prolotherapy injections have not been proven to be as effective as or more effective than placebo injections. Additional randomized studies are needed to evaluate the efficacy of prolotherapy for joint or ligament instability. The evidence in the peer-reviewed, published scientific literature is insufficient to support the use of prolotherapy for any therapeutic indication.
Although proponents have advocated the use of prolotherapy for a range of indications, relatively few clinical uses have been studied systematically or published in the peer-reviewed literature. Results of the most recent systematic reviews are inconclusive for demonstrating the effectiveness of prolotherapy for treatment of musculoskeletal pain, and new evidence from case series would not alter these conclusions. The majority of published experimental studies have included conservative therapy with prolotherapy for relief of chronic low back pain, and to a lesser extent, osteoarthritis of the knee with varying results. Sample sizes have been insufficient on which to base national policy decisions.
These reviews are particularly useful as sources of references to studies examining prolotherapy. Reviewing the studies cited in these summaries, it is clear that the majority of positive results stem from trials without proper controls: randomization of subjects, placebo or no treatment control groups, proper blinding, etc. And because the agents used and other aspects of the treatments are inconsistent between studies, it is difficult to compare studies or to generalize from one study to a different prolotherapy treatment approach, and it is impossible to combine small studies in meta-analyses that might have greater power to determine if the treatment is truly effective.
It is often the case that smaller and less rigorously designed studies tend to have positive conclusions. As the quality and quantity of research improves, many of these results turn out not to be trustworthy. Prolotherapy in humans is currently supported by a limited quantity of low-quality research. Until stronger evidence accumulates, if it does, the practice should be viewed as an experimental therapy of unknown benefits and reserved for circumstances in which conservative management and established conventional therapies are unsuccessful.
As is also often the case, there are absolutely no rigorous, controlled clinical studies of prolotherapy in dogs and cats. The use of this approach in pets is based entirely on anecdotal evidence, which is highly unreliable, and on extrapolation from human medicine, in which the practice has little supportive evidence and is not widely accepted as a legitimate therapy. Prolotherapy might be justifiable as an experimental intervention in cases in which better studied therapies with more established physiological rationales have failed or cannot be used, but it should not be marketed with strong claims of “proven” benefits, and it should not be used in lieu of conventional treatment.
Is it Safe?
Any injection into a joint must be done with careful attention to technique to avoid introducing infection or damaging tissues, and sedation may be necessary for such treatments, so these are risks associated with but not unique to prolotherapy treatment of joints. In humans, there have been limited efforts to assess the risks of prolotherapy. One survey of practitioners of this treatment has been reported.(7)
Dagenias et al. (2006) studied the effects and adverse events related to prolotherapy for back and neck pain. The authors conducted a practitioner postal survey evaluating prolotherapy for back and neck pain in the United States and Canada. Surveys were mailed to members of the American Academy of Orthopaedic Medicine (AAOM) and the American College of Osteopathic Pain Management and Sclerotherapy (ACOPMS), both closely affiliated with prolotherapy conference, in addition to nonmembers attending the 2004 AAOM annual conference. A 50% response rate was obtained. The authors published that the side effect with the highest estimated median prevalence was temporary post-injection pain (70%), stiffness (25%), bruising (5%), and temporary numbness (1%). The most commonly reported adverse events (total of 472) were spinal headache (n=164), pneumothorax (n=123), temporary systemic reactions (n=73) and nerve damage (n=54). A total of 69 adverse events required hospitalization, and five resulted in permanent injury, secondary to nerve damage. Almost 98% of the respondents held MD or DO degrees, and 83% were board certified in related disciplines. The authors concluded that to better assess the true risk of adverse events related to prolotherapy, further well-designed clinical trials that are designed to over-come bias inherent to practitioner surveys are needed.
Overall, prolotherapy likely presents a low to moderate risk depending on the agents used and the skill and experience of the practitioner.
Prolotherapy is a purported treatment for connective tissue and joint pain and disability. It involves injecting substances which induce inflammation and other chemical and cellular reactions into affected tissues. These reactions are theorized to relieve pain and improve function. The logic of this theory is questionable, and no clear mechanism for beneficial effects from prolotherapy has been described, but it is possible that the theory could be valid.
The clinical research on prolotherapy in humans is generally of low quality and results have been mixed. There is great variation in the techniques used by different investigators, so it is difficult to compare or generalize between studies.
There is virtually no controlled research investigating prolotherapy in companion animals, and all claims made for safety and efficacy in these species are based solely on anecdotal evidence.
The use of proltherapy in pets should be viewed as experimental with unknown risks and benefits. Such treatments should be reserved for patients that have significant symptoms that have failed to respond or cannot be treated by conventional means.
1. Banks, A.R. A rationale for prolotherapy. Journal of Orthopaedic Medicine. 1991;13(3). Accessed at http://www.prolotherapy.com/articles/banks.htm July 13, 2011.
2. Robinson, NG. Prolotherapy for pain. 2007. Accessed at http://csuvets.colostate.edu/pain/Articlespdf/Prolotherapy%20for%20Pain.pdf July 13, 2011.
3. Dagenais S, Yelland MJ, Del Mar C, Schoene ML. Prolotherapy injections for chronic low-back pain. Cochrane Database of Systematic Reviews 2007, Issue 2.
4. Aetna. Clinical Policy Bulletin: Prolotherapy. Last revised May 24, 2011. Accessed at http://www.aetna.com/cpb/medical/data/200_299/0207.html July 13, 2011.
5. Cigna HealthCare Coverage Position: Prolotherapy. Last revised December 15, 2006. Accessed at http://stage.cigna.com/health/provider/medical/procedural/coverage_positions/medical/mm_0006_coveragepositioncriteria_prolotherapy.pdf July 13, 2011.
6. Adams E. Bibliography: Prolotherapy for musculoskeletal pain. Boston, MA: Veterans Administration Technology Assessment Program (VATAP); April 2008.
7. Dagenais S, Ogunseitan O, Haldeman S, Wooley JR, Newcomb RL. Side effects and adverse events related to intraligamentous injection of sclerosing solutions (prolotherapy) for back and neck pain: A survey of practitioners. Arch Phys Med Rehabil. 2006 Jul;87(7):909-13.
ConsumerLab lists two other double-blind placebo controlled studies of prolotherapy. I tend to think of them as a fairly reliable, objective site for information.
I have a dog with severe bilateral HD, who’s too old for a hip replacement, and suffering – I’m trying any and everything I can ‘cuz she has so much life otherwise
would love your insights on these:
Reeves KD, Hassanein K. Randomized prospective double-blind placebo-controlled study of dextrose prolotherapy for knee osteoarthritis with or without ACL laxity. Altern Ther Health Med. 2000;6:68-70,72-74,77-80.
Reeves KD, Hassanein K. Randomized, prospective, placebo-controlled double-blind study of dextrose prolotherapy for osteoarthritic thumb and finger (DIP, PIP, and trapeziometacarpal) joints: evidence of clinical efficacy. J Altern Complement Med. 2000;6:311-320.
Thanks for pointing out these two articles. They have been evaluated previously in a systematic review of prolotheray which concluded they did provide some support for the possible benefits of the practice, though like all research they had limitations. In terms of external sources of potential bias, one has to recognize they were both published in alternative edicine journals, which tend to publish positive studies of alternative therapies at a higher rate than negative studies, which can create a misleading oveall impression (as happens also in the conventional literature, of course).
The knee study is available here. It’s a pretty well-done study, but there are a couple of caveats. There was statistically significant imporvement in various measures of pain for both the placebo and the dextrose solution, and the differences were very small and of questionable clinical significance. For some measures of pain, the placebo group actually improved more than the dextrose group, which always suggests there may be a problem. The overall difference between the groups was reported as statistically significant (though the details of the comparison were not reported in the paper), but again the inconsistency in the results makes me wonder a bit.
The other study can be read here. This two showed some differences between treatment and placebo, but these were not large and not always greater than chance. Also, this study involved only 13 subjects, which is too few to generate reliable conclusions.
The lead author of these studies has posted a list of prolotherapy research studies on his web site. Checking both there and on Pubmed, I can find no controlled clinical trials of prolotherapy for arthritis in the 14 years since these two were published. It’s a bit suspicious that no one appears to be following up on this idea if it is as promising and effective as its promoters claim.
I certainly understand your desire to do whatever you can for your dog. Prolotherapy is one of those treatments about which we really can’t say with any certainty if it helps. Unfortunately, that also means we can’t be certain it isn’t harmful since the data on that question is just as incomplete as the efficacy data. It always worries me a bit that in our desire to do something in such situations we can easily forget we might be making things worse rather than better.
Whatever you choose to do, I hope it works out well for you and your dog.
Here is a more recent article that I would appreciate your comments on: holisticandorganixpetshoppe
I would be very skeptical of anything published by insurance companies such as Aetna and Cigna. By attempting to debunk this procedure, they can continue to refuse to cover them due to their being “experimental.”
Finally, how do you feel about the use of hyaluronic acid injections in dogs for conditions such as DJD?
The site you link to contains no evidence for prolotherapy, simply anecdotes from a believer, which is not a reliable kind of evidence. The person at that site also makes all sorts of claims not based on any scientific evdience, including blaming grains in the diet and recommending raw diets, both of which are unproven and most likely nonsense. So it is hardly a reliable site.
As for the insurance companies, there is no question they have a vested interest in not paying for therapies if they can show these therapies are ineffective. However, the comapnies did not conduct or pay for the research I have linked to, they simply collected a number of papers in one place. The systematic reviewes which one can read at those sites are independant of the insurance companies, so the bias the companies undoubtedly has did not influence the scientific evidence these links lead to. If you wish to argue in favor of prolotherapy, it is not sufficient to cite other believers and question the motives of critics. Real scientific evidence is what is needed to decide what, if any, value this treatment has.
I was researching this treatment and came across the below, newer studies.
Can you comment, please?
The problem with these papers is that they are uncontrolled and unblinded. A key element in evaluating whether or not a therapy works is comparing it to either a placebo or an established therapy in circumstances in which no one knows which treatment they are getting. This has revolutionized medicine by eliminating many treatments that seemed to work but failed this kind of controlled test.
So these papers basically involved people who believe prolotherapy is useful knowingly giving prolotherapy to people who knew they were getting it and then deciding whether it seemed to work. Such results are helpful, but they can never tell us for certain if a treatment works or doesn’t work. If the outcome is completely objective (say death), then lack of blinding may only be a slight problem. When the outcome is entirely subjective (perception of discomfort), as in these trials, lack of blinding is a much bigger concern.
So while these add a bit of weak evidence to the subject, they don’t really change the balance. And, of course, they don’t apply directly to pets, only humans.
Thanks for the links!
Thank you, Skeptvet. Your research and analysis seem very helpful. Of course, we realize that any treatment that wanders off the path of patentable prescriptions or surgery will experience little financial support for the research necessary to establish efficacy/safety. Still, I appreciate your posting this article. You have helped me to understand that prolotherapy is likely another example of over-treatment where the risk of causing more harm makes it an undesirable therapy.
What do you think of this 2013 study from researchers at the University of Wisconsin http://www.ncbi.nlm.nih.gov/m/pubmed/23690322/
which does include blind protocols? As long as there are no significant side effects, and the surgery options are not that great, I’m considering trying it for my dog’s rear knee
In humans, a few recent studies have shown some benefits for knee arthritis. Here is a brief summary from the Science-based Medicine blog which includes the study you mention and a couple of others:
I tend to agree that these recent results are encouraging, though not definitive. I would also emphasize that humans and dogs are different in many ways, and since there are no similar trials in dogs, we’re guessing about whether or not the results, both safety and efficacy, will be the same. So I don’t think it is unreasonable to consider doing this. But I do wonder what you mean by “the surgery options are not that great?” If you are talking about a cruciate ligament rupture, surgery has been proven to be effective and so is a much better option. For dogs with chronic arthritis who don’t do well on systemic medications, prolotherapy is something to consider.
I am a physician with over 30 years of experience . I have been using Prolotherapy for the management of pain around peripheral joint and spinal joints with an average of 80% positive outcome. That is 8 out of ten patients will have significant reduction in pain and an improvement in function. After a 3-8 treatments to an area, the results can last for years. This depending upon the activity level. A Runner for example, may need treatment every 6 months.
In the current environment, it cost millions of dollars to bring a drug to market. It is unrealistic to expect anyone to spend the money necessary to research prolotherapy that would meet Cockrane criteria. Most of the medical literature that supports the use of prolotherapy is performed by physician’s, like myself with a small number of subjects. A satisfactory study would require an infusion of much money to substantiate the use of the injectables that could not be patented, and thus never pay back the cost of the research investigation. We will have to be happy with anecdotal reports like the ones you discuss above.
You reference a postal survey of physician’s concerning morbidity of the prolotherapy procedure. First off, how does a postal survey where only 50% responded, meet Cochrane criteria? The inclusion of orthopedists in this group is also misleading, as they are surgically trained. In experience at teaching and community hospitals and outpatient clinics, they do not perform prolotherapy. I have to assume that their reporting is based on their training as well as a financial incentive. Prolotherapy performed by myself and other physician’s I have known, has a much lower morbidity than surgery any day. Pneumothorax or nerve injury is probably a result of operators performing prolotherapy with minimal experience of the procedure and limited knowledge of surface anatomy.
Are these numbers from published controlled studies or simply your clinical impressions?
I may be an “N” of 1, but I’ve used prolotherapy for 5 years for SI joint pain. It has allowed me to be active and mobile- 5 mile walks, hill climbing and bicycling. I’ve had 1 prolotherapy injection in the basal joint in my thumb and it eliminated pain there due to arthritis. My experience with prolotherapy is that it works very well for about 4 to 6 months and then wears off. So I’ve maintained a strong level of fitness with these injections, without using steroid injections or oral pain meds. I’m looking to try it on my arthritic dog.
I understand your reasoning, but unfortunately anecdotes and personal experiences don’t reliably predict what will be safe and effective medicine for our pets. We are not dogs, our knees are not dog knees, and only controlled research can actually tell us whether something like this will work or not.
Here’s some more detailed discussion of this problem.
Why Anecdotes Can’t Be Trusted